You are on page 1of 52

Duke Law Journal

VOLUME 42

OCTOBER 1992

NUMBER 1

LEGAL COMPLEXITY: SOME CAUSES,
CONSEQUENCES, AND CURES
PETER

H. SCHUCK

,

INTRODUCTION

Simplicity is a compelling virtue. Poets prais6 it.' Artists pursue it.' Moralists preach it? Monastics live it.' Even today-especially today, as simplicity becomes more elusive-we more ordinary folk pay it at least verbal homage.

Scholars' views of simplicity are, well, more complex. On the
one hand, simplicity cuts across our intellectual grain. We have a
strong taste for complexity; indeed complexity amounts to a craft

value-a point to which I shall return. On the other hand, scholars
embrace simplicity (at least in its more sophisticated forms) out of

professional ideology and custom. Recognizing that the cost of
information is a crucial constraint on greater complexity, scholars

t Simeon E. Baldwin Professor of Law, Yale University. An earlier, abbreviated
version of this Article was delivered as the 1992 Brainerd Currie Lecture at the Duke
University School of Law on .March 30, 1992. I wish to express my gratitude to the
Currie family and the School for establishing the Lecture and inviting me to deliver it. I
also wish to acknowledge the useful comments of members of the Duke faculty, my Yale
colleagues who participated in a faculty workshop on the paper, and Professors Charles
Reich and Marc Galanter. Bill Malley and Margo Schlanger helped prepare the footnotes.
1. "And, as the greatest only are, [i]n his simplicity sublime." ALFRED LORD TENNYSON, ODE ON THE DEATH OF THE DUKE OF WELLINGTON 6 (1852).
2. "Art, it seems to me, should simplify. That, indeed, is very nearly the whole of
the higher artistic process . . . ." WILLA CATHER, ON WRITING 102 (1949).
3. "Embrace simplicity .... ." LAO-TzU, THE WAY OF LAO-ZU (Wing-tsit Chan
trans.), reprinted in JOHN BARTLETT, FAMILIAR QUOTATIONS 64 (Emily M. Beck ed.,

15th ed. 1980).
4.

"Our life is frittered away by detail ....

Simplify, simplify."

THOREAU, WALDEN 91 (J. Lyndon Shanley ed., 1971) (1854).

HENRY D.

DUKE LAW JOURNAL

[Vol. 42:1

view simplification as a necessary concession to our limited knowledge. This insight yields many fruitful implications.' Scholars perceive another of simplicity's virtues, this one at the meta-scientific
level. There, the principle of Ockham's Razor asserts that simpler
theories possess greater methodological, epistemological, and aesthetic elegance and value.6
I propose to explore complexity, simplicity's obverse-and
more specifically, legal complexity. Of course, any attempt to analyze this subject systematically is fraught with difficulties. After all,
legal complexity is hard to define, much less measure. It arises in
a multitude of different contexts that may have little in common,
contexts in which its meanings are likely to be different and its
consequences not readily traceable. The task of distinguishing
when it is a cause of some other condition from when it is an
effect of some other cause, always difficult, is especially so because of complexity's frequent association with cognate concepts
such as uncertainty. Even when its effects can be isolated, evaluating them is not a straightforward matter. In short, those who
attempt to generalize about, and hence to simplify, a subject this
subtle and intricate should reflect upon the words of the
curmudgeonly Henry Adams, who denounced simplicity as "the
most deceitful mistress that ever betrayed man."7
For all its pitfalls, however, the effort to analyze legal complexity seems a worthy enterprise. This is certainly not because
legal complexity is a new subject, In fact, it is an ancient con-

5. Herbert Simon, for example, noted the irrationality of complex, formally rational
decisionmaking in organizations, and developed a fundamental distinction between optimizing and "satisficing" behavior. HERBERT SIMON, ADMINISTRATIVE BEHAVIOR xxviiixxxi (3d ed. 1976). In the political sphere, Charles Lindblom described, and sought to
justify, incremental behavior-what he called "partisan mutual adjustment"-in terms of
the need to reduce information and other decisionmaking costs. See CHARLES E.
LINDBLOM, THE INTELLIGENCE OF DEMOCRACY: DECISION MAKING THROUGH MUTUAL
ADJUSTMENT 21-84 (1965); Charles E. Lindblom, The Science of "Muddling Through", 19
PUB. ADMIN. REV. 79 (1959). Ronald Coase analyzed how various kinds of complexity
affect firms' structure and their pursuit of efficiency. See Ronald H. Coase, The Nature of
the Firm, 4 ECONOMICA N.S. 386 (1937), reprinted in THE NATURE OF THE FIRM (Oliver
E. Williamson & Sidney G. Winter eds., 1991); Ronald H. Coase, The Problem of Social
Cost, 3 J.L. & ECON. 1 (1960) [hereinafter Coase, The Problem of Social Cost]. Oliver
Williamson extended Coase's insights in a variety of ways. See, e.g., OLIVER WILLIAMSON, MARKETS AND HIERARCHIES: ANALYSIS AND ANTITRUST IMPLICATIONS (1975).
6. For an elaborate historical development of this point, see THOMAS S. KUHN,
THE STRUCTURE OF SCIENTIFIC REVOLUTIONS (1962).
7.

HENRY ADAMS, THE EDUCATION OF HENRY ADAMS 441 (1918).

1992]

LEGAL COMPLEXITY

cern, 8 one that has continued to obsess lawyers, policymakers,
social critics, editorial writers, and men and women in the street,
who almost invariably condemn legal complexity in the most caustic terms.9 Yet despite the cascade of public and professional criticism, consideration of legal complexity in the academy is still in
an embryonic stage; the subject remains more narrowly conceived
than its importance warrants.
I define a legal system as complex to the extent that its rules,
processes, institutions, and supporting culture possess four features:
density, technicality, differentiation, and indeterminacy or uncertainty. I hasten to concede that one might well conceive "and define it in other terms. An alternative and potentially useful definition, for example, might center on whether the people who are
subjected to a legal rule, process, or institution perceive and experience it as complex, instead of on the objective qualities that a
more detached viewer might ascribe to it (as my definition
does)." Still, I suspect that the four qualities on which I focus
capture most, if not all, of what people would mean if they were
to think about legal complexity.
Density and technicality are chiefly aspects of the system's
rules. Dense rules are numerous and encompassing. They occupy a
large portion of the relevant policy space and seek to control a
broad range of conduct, which causes them to collide and conflict
with their animating policies with some frequency. An example of
a dense legal regime is that governing pension administration,"

8. As Roscoe Pound noted in 1906, "[d]issatisfa ction with the administration of
justice is as old as law." Roscoe Pound, The Causes of Popular Dissatisfaction with the
Administration of Justice, 40 AM. L. REv. 729, 729 (1906), reprinted in 8 BAYLOR L.
REV. 1, 1 (1956). Several of the problems that Pound discussed, such as the law's procedural focus and the "sporting theory of justice," are elements of legal complexity, although he did not use that precise phrase.
9. See, e.g., WALTER K. OLSON,.THE LITIGATION EXPLOSION (1991).
10. Even parties subject to the same rule and engaged in the same transaction-the
la~v
governing an insurance contract, for example-might well perceive its level of complexity very differently, depending on their information and other resources and their
stakes in and experience with the rule. Another aspect of the difference between these
"objective" and "subjective" definitions of complexity is nicely illustrated by the Tarasoff
litigation, discussed infra note 36. There, the therapists understood the court's rule to
require them to warn those in danger, even though the rule' in fact imposed only a general duty of care, which might or might not include a warning-a confusion facilitated by
the court's alteration of its original decision. I am grateful to Marc Galanter for raising
this point and these examples.
11. 29 U.S.C. §§ 1001-1461 (1988).

DUKE LAW JOURNAL

[Vol. 42:1

which cuts across and seeks to integrate a wide variety of legal
specialties.
Technical rules require special sophistication or expertise on
the part of those who wish to understand and apply them. Techni.cality is a function of the fineness of the distinctions a rule makes,
the specialized terminology it employs, and the refined substantive
judgments it requires. The Internal Revenue Code12 is probably
the leading example of technical rules.
A legal system is institutionally differentiated insofar as it
contains a number of decision structures that draw upon different
sources. of legitimacy, possess different kinds of organizational
intelligence, and employ different decision processes for creating,
elaborating, and applying the rules. Product safety, for example, is
institutionally differentiated in that it is governed by statutory
provisions, regulatory standards promulgated by several different
agencies and private technical organizations, tort litigation, and
common law contract principles. 3
The final feature-indeterminacy-is a quality both of rules
and of legal processes and institutions. Indeterminate rules, processes, and institutions are usually open-textured,1 flexible, multifactored, and fluid. The familiar reasonableness standard in tort
law is an example of an indeterminate rule. Turning on diverse
mixtures of fact and policy, indeterminate rules tend to be costly
to apply and their outcomes are often hard to predict.
Indeterminacy's relation to legal complexity is itself complex. Ironically, rules and institutions that are designed to reduce the law's
indeterminacy may actually increase it, due to the cumulative effect of their density, technicality, and differentiation." Indeterminacy, then, may be a consequence, as well as a defining feature, of
complexity.
Although I have been discussing complexity as if it were a
fixed essence, it is of course nothing of the sort. Indeed, I have

12. 26 U.S.C. (1988).
13. See generally JERRY MASHAW & DAVID HARFsT, THE STRUGGLE FOR AUTO
SAFETY (1990); see also Competitive Enter. Inst. v. National Highway Traffic Safety
Admin., 956 F.2d 321 (D.C. Cir. 1992) (dispute between automobile lobby and NHTSA
over fuel economy standards).
14.

This concept is developed in H.L.A. HART, THE CONcEPT OF LAW 124-32

(1961).
15. See Boris I. Bittker, Tax Reform and Tax Simplification, 29 U. MIAMI L. REV.
1, 2 (1974).

1992]

LEGAL COMPLEXITY

just defined it as a composite of four variables. As such, it can

only be located on a continuum ranging from extreme simplicity at
one end to extreme complexity at the other. Thus a legal rule,
process, or. institution is neither simple nor complex; it is only
more or less simple or complex compared to some other actual or
ideal one. Complexity is multi-dimensional, and its dimensions
cannot easily be measured, much less weighted. Hence it is neither
possible nor desirable to attempt to classify something precisely as
simple or complex. I shall therefore use the terms in a loose,
relativistic sense, always assuming that other things remain equal.
As long as my general analysis persuades, little should turn upon
how any particular legal phenomenon is classified.16
Even so, some examples may help to sharpen the categories a
bit. At the extremes, the rule that the President of the United
States must be at least 35 years old17 is simple, whereas the rule
against perpetuities"5 is complex. Unlike the presidential age rule,
the rule against perpetuities is embedded in a dense forest of
norms, is notoriously indeterminate and highly technical (neither
rule is institutionally differentiated). Moving away from the poles,
tort law's reasonableness standard is moderately complex. Although neither dense nor technical, it is notably indeterminate,
involving interaction among judge, jury, private groups, and often
a legislature or agency. Precisely because contract rules generally
seek to facilitate private choices, they tend to be fairly simple. Tax
law, while including many bright-line rules, is quite complex; its

16. How, for example, should a system of scheduled damages for pain and suffering
be classified relative to the existing system of largely indeterminate, jury-prescribed
awards? In my discussion of these systems in Part IV, I treat the former as simpler than
the latter. I do so because the former is relatively determinate and institutionally undifferentiated (the jury has much less to decide), and requires a less technical set of doctrines for controlling the jury. On the other hand, a schedule system could be quite
complex, depending on how it is designed. Classification, then, is manifestly a matter of
judgment.
17. U.S. CONST. art. II, § 1, cl. 4.
18. See Lucas v. Hamm, 364 P.2d 685 (Cal. 1961) (holding that rule against perpetuities so complicated that attorney of ordinary skill who misunderstands it cannot be held
liable for malpractice), cert. denied, 368 U.S. 987 (1962). The modification of this rule
proposed in the Uniform Statutory Rule Against Perpetuities, adopted in a number of
jurisdictions, exemplifies one approach to legal simplification-here, through a savings
clause principle of wait-and-see. See UNIFORM STATUTORY RULE AGAINST PERPEBurTIES
prefatory note, reprinted in JOHN H. LANGBEIN & LAWRENCE W. WAGGONER, SELECTED STATUTES ON TRUSTS AND ESTATES 674, 674 (1992).

Reducing Law's Uncertainty and Complexity. costs. CmH. which I develop and defend in turn in Parts I through IV. including legal uncertainty. 257. Chaos. accessibility. REV. and congruency. sources cited supra note 5. 76 CORNELL L. judicial decisions. L. Werner Z. Johnston. subject matter. . Its forms. e. Still. My argument consists of four linked claims. has consequences that go well beyond transaction costs. The Federal Sentencing Guidelines: A Dismal Failure 6 (remarks at the University of Chicago Law School. Posner.g. technical. The Failure of Sentencing Guidelines: A Plea for Less Aggregation. Colin S. and that this is problematic for a system of justice-are not. REV. Diver. Of course. 67 (1983). and remedial structure of legal norms in a variety of areas. An Economic Analysis of Legal Rulemaking. and rule form. 42:1 rules are dense. 58 U. Issac Ehrlich & Richard A.' These analyses are valuable and form the basis for much that follows.DUKE LAW JOURNAL [Vol. I think. especially litigation incentives. and (2) certain of its sources. See.J. L. and possible reform raise important social issues that need to be explored in greater depth and breadth. see Anthony D'Amato. 71 CAL. Jos6 A. LEGAL STUD. 1992) (on file with author). 341 (1991). REV. 901 (1991). one is struck by their narrowness. Some claim-perhaps hyperbolically.g. Some scholars have analyzed the optimal level of rule precision or determinacy. I hope in Part I to satisfy any doubters by showing that this trend appears in the forms. January 15. Many scholars have noted the transaction costs of complex rules. and elaborated through a differentiated system of agencies and tribunals. Alschuler. On legal uncertainty. nevertheless.. goals. Hirsch. a genre summarized and extended in Jason S. Legal complexity. interpretation. and the Torts Process: An Economic Analysis of Legal Form." For all the broad-gauged interest in legal complexity. 3 J. Legal Uncertainty. Cabranes. e. legal scholars have largely confined their analyses of this phenomenon to two aspects: (1) its transaction costs. 1 (1983). whereas Colin Diver defines it in terms of transparency. The first two claims-that legal complexity is increasing. Uncertainty. The Optimal Precision of Administrative Rules. 20. see also Albert W. 1233 (1974).. particularly controversial. I can no more demonstrate the first (growing complexity) by merely citing examples than I can prove increased regulatory burdens by counting pages in the Federal Register (although many have purported to do so). limits. See. 65. after all. 261 (1974). and has sources besides litigation incentives and judicial lawmaking. 93 YALE L. certainly antagonistically-that the Federal Sentencing Guidelines are approaching the level of complexity of the Tax Code. Ehrlich and Posner define precision in terms of the number and simplicity of the facts to which legal consequences attach. Another group has explored how rule forms evolve through litigation. structural causes. REV. 19. 21 UCLA L.

at 10. 23. infra. it deflects attention from two other kinds of complexity-induced problems. perhaps because they are less palpable and involve "softer" responses of a psychological and emotional nature. AGENT ORANGE ON TRIAL: MASS TOXIC DISASTERS IN THE COURTS (enlarged ed. it ignores another important set of causes. . POL'Y 541.. & PUB. this simplification is more often de facto than de jure. legal complexity. point to the administrative and transaction costs that complexity generates.21 Although this point is correct as far as it goes. My final claim. Although this approach seems correct as far as it goes. These have received much less attention. SCHUCK. e. Scholars analyzing legal complexity have largely defined it in terms of uncertainty and other transaction costs. My analysis of the political economy of legal complexity in Part III may help to explain this interesting lag among academics. See. Peter H. litigators. is that the growing costs of legal complexity generate powerful behavioral responses in the direction of legal simplification 23-first among private decisionmakers. including mass toxic torts. See. Numerous commentators. I seek to remedy this neglect in Part II. administrators. 22. Complex litigation. 24. 15 HARV. that complexity is problematic for a system of justice. 1987). and legal scholars-to choose more. broadly defined. My third claim speaks to the causes of legal complexity and is therefore more controversial. As discussed in Part III. supra note 20. 1992).g. I 21. which I call the "political economy of complexity. and craft values that drive the dominant producers and rationalizers of law-here I refer not only to judges but also to legislators and their staffs. among legal scholars24 -and that these responses can be used not only to define complexity's limits but also to deduce general principles for evaluating particular reform proposals. D'Amato. presented in Part IV. The Worst Should Go First: Deferral Registries in Asbestos Litigation. career ambitions. RICHARD A. PETER H. rather than less. Hirsch. e. 553-68 (1992).. sources cited supra notes 20-21. supra note 20. then among public officials. POSNER. J.L.19921 LEGAL COMPLEXITY My second claim. at 1241-45. Schuck. will likewise command widespread agreement.g. is perhaps the most widely discussed example. and relatively late in the game. ECONOMIC ANALYSIS OF LAW 542-47 (4th ed." As discussed in Part III. this consists of -the professional skills. much complexity originates in the ways in which litigation incentives and adjudication affect the form of rules. which I call governance and delegitimation costs." According to them. for example.

Joskow & Alvin K. even more certain. And to write a Comment upon any Law is a capital Crime. GULLIVER'S TRAVELS 130 (Scholars' Facsimiles & Reprints 1976) (1726). complexity should not be seen as a symbol of what is wrong with American law more generally. As Henry Adams's epigrams' and Jerry Brown's flat tax proposal remind us. the reverse can also be true: simple language can be a source of complexity. we end up with more complexity than we need (by some more objective account).. rationalizers. By the same token. their Precedents are so few. . JONATHAN SwiFT. It is purposeful. forms of social controlY The question always is: All things considered. 89 YALE L. which consists only in two and twenty. to suggest some possible cures. Because the political economy of lawmaking presses the law toward greater complexity with little regard to whether any particular complexity is worth its costs. This caveat-that simpler law is not always better law-is important because I shall say a great deal about complexity's costs but little about its benefits. Paul L. Indeed complexity is both a weakness and a strength. are more limited and theoretical: to enrich the discourse on complexity's causes and consequences.DUKE LAW JOURNAL [Vol. An important caveat going to my normative argument is in order." Legal complexity sometimes produces fairer. 213 (1979). See. more efficient. which we can only hope to redress by urging the producers. I wish to be clear about this bias. 42:1 conclude in Part V by identifying some specific simplification options that should be seriously considered. of course. But indeed. then. at 1-2. that they have little Reason to boast of any extraordinary Skill in either. wherein those People are not mercurial enough to discover above one Interpretation. more refined. lest it be misunderstood. As to the Decision of civil Causes.g. indeed it connects with my central point. See supra text accompanying note 7. 27. supra note 15. and I do not seek to answer it for any particular legal phenomenon. . few of them extend even to that Length. e. As Bittker points out. Bittker. Klevorick. The absurdity of a thoroughgoing legal simplicity was captured by Swift in his description of Brobdingnag's legal system: No Law of that Country must exceed in Words the number of Letters in their Alphabet.J. Bittker. simpler is not always better. at 2. 26. This constitutes a kind of structural imbalance. A Framework for Analyzing Predatory Pricing Policy. They are expressed in the most plain and simple Terms. or Proceedings against Criminals. are the benefits of a given level of complexity worth its costs? There is no general answer to this question. supra note 15. My goals. and administra- 25. and to derive some criteria for knowing when to pursue simpler law.

g. a feature that it appears to share with other systems. This may reflect a more secular.LEGAL COMPLEXITY 1992] tors of law to keep simplicity's more elusive virtues and complexity's corresponding vices constantly in mind. Even here. 40 STAN. supra note 20 at 346. 1 Rules are not always simple. Consider the deep structural differences between tort law (decentralized. acontextual. The Form of Legal Norms Many legal norms evolved from relatively precise. contextual. These two types of forms are sometimes referred to as rules and standards. 578-79 (1988). See Johnston. 31. and hard-edged forms to relatively ambiguous. determinate. this tendency is certainly not inexorable. 1685. and influences the nature of social relationships. values. Form and Substance in Private Law Adjudication. Ehrlich & Posner." Carol M. THE ENTROPY LAW AND THE ECONOMIC PROCESS (1971). 89 HARv. TALCOTT PARSONS. goals. subject matter. and standards are not always complex. physical and social. administrative law (process-oriented). interpretation. Moreover. patent law (technology-driven." Yet for all these qualifications. antipodal types. THE GROWTH IN LEGAL COMPLEXITY The legal system as a whole exhibits a marked tendency to become more complex. They express distinctive ways in which the law views. 9 Nor does complexity grow at the same rate and at the same time in all areas of the law. perhaps even universal. common law). . and remedial infrastructure of legal norms. but these formal categories do describe paradigmatic. REv. however. A. Further. 29. Carol Rose calls them "crystals" and "mud. Given such differences. indeterminate. Duncan Kennedy.. Rose. argues about. REV. property law (transaction-oriented)." In the case of the law. as the replacement of writ pleading with much simpler code pleading and the replacement of tort law in the workplace with workers' compensation systems demonstrate. L. one is also struck by the tendehcy of the simpler substitutes to grow more complex over time. 577. See. One can discern it throughout the legal system-in the form. and constitutional law (structure-oriented). I. 30. dynamic affecting such systems. the similarity of the rhythm and character of change in these fields is unexpected. THE SOCIAL SYSTEM (1951). these formal types affect 28. supra note 20. L. statutory). See generally NICHOLAs GEORGESCU-ROEGEN. the strength and generality of this trend are striking. at 258. particular doctrines may oscillate between simpler and more complex forms. Crystals and Mud in Property Law. e. 1687-88 (1976). and open-ended ones.

in large part because of the greater certainty that rules were thought to afford. 33. This study.. see also infra text accompanying notes 164-65. Peter H. judicial review. A generation of scholars. and indeterminate than if the legislature had simply promulgated a rule itself. See generally KENNETH C. 42:1 substance.4%. and procedural requirements are examples-the pervasive delegation of discretion to agencies (and to courts) means that the resulting legal regime is almost certain to become more dense. 1014-15 (in 1984-1985.J. technical. and reformers urged that agencies make policy through rulemaking rather than the traditional case-by-case adjudication. Don Elliott and I found that during the last twenty years rulemaking made only minor inroads. . See id. they constituted only 6. even given 'the risk that it will sometimes be abused. 984. There are often compelling public interest reasons to confer discretion. the high point in absolute terms for rulemakings during the study period.34 and the exercise of discretion in turn generates more legal complexity. by counting circuit court cases involving such proceedings.5% of cases coded as rulemakings.3 The movement toward administrative rulemaking is intimately connected to the increased resort to administrative discretion as a dominant legal modality. should be seen as one more consequence of doing so. 32. Schuck & E. however. rulemakings constituted only 9. although animated in part by a quest for greater certainty. measured rulemakings only indirectly. officials. as the history of administrative law demonstrates. DISCRETIONARY JUSTICE: A PRELIMINARY INQUIRY (1969). the share of rulemaking declined between 1975 and 1984-1985).DUKE LAW JOURNAL [Vol. ratemakings. At the "meta-law" level. public participation. legislative oversight. Greater legal complexity. To the Chevron Station: An Empirical Study of Federal Administrative Law. But in a longitudinal study of administrative law court decisions.' Indeed we found a retreat from rulemaking back to adjudication in several important policy sectors. 1990-DUKE L. shaping both particular legal outcomes and the balance among competing social ideals and practices. has only produced more open-ended complexity. institutionally differentiated. A movement from rules to more complex standards has been evident for some time. 34. Efforts to achieve greater certainty have generally failed. it should be noted. Donald Elliott. excluding agencies that did not use rulemaking at all. DAVIS. the evolution of legal norms from common law principles to statutes to administrative regulation. or adjudications. Although the legal system employs many techniques to constrain and guide discretion-statutory language.

Hedlund v.g..W. In the areas cited in the text. L.g. 1989) (extending duty to any foreseeable victim. 443 P. impact. Keaton. of Cal. Consolidated Rail Corp. Givelber et al.2d 561 (Cal. of Cal.2d 230. County of Maricopa. 443. 1128 (Ariz. the no-duty rule. 775 P. more general duty that has spawned much confusion. See Daniel J. 6 Another example from the same court is found in Rowland v.2d 41. When the Exception Becomes the Rule: Regulatory Equity and the Formulation of Energy Policy Through an Exceptions Process. E. 236-37 (Wash.2d 553 (Cal. Johnston. 669 P. 671 P... although child was not defendant's patient). cases cited in RICH- ARD EPTIN ET AL. Compare Tarasoff v. 37." An especially striking example is found in the well-known Tarasoff case. even more indeterminate set of principles. Christian.. 495 A. State. 36. 1984 DUKE L. 465-68 (showing confusion in therapists' understanding of difference between Tarasoff ls duty to warn and Tarasoff I1s more general duty).2d 1122.1992] LEGAL COMPLEXITY Even tort law.J. 1976) with Tarasoff v. e. CASES AND MATERIALS ON TORTS 1055-56 (4th ed. For an example of stretching in administrative law. has become more complex. e.J. for example. but also as reducing the rule's efficiency.. 46-47 (Cal. judicial recognition of relatively open-ended cancerpho- 35. 240 N. Farwell v. 551 P. 222 (Mih. . after imposing a specific duty to warn foreseeable victims. 1985). Subsequent cases have glossed Tarasoff II with a complex. 1974) (Tarasoff I). a vaguer. 529 P. finding duty extended to child who suffered emotional harm while seated next to mother who was attacked.g. not just target of specific threat). courts applying relatively clear rules stretched their meanings in order to find: a special relationship as the predicate to imposing a legal duty. the impact rule limiting emotional distress claims. withdrawn. see Peter H. People Express Airlines v.2d 334 (Cal. the rule restricting recovery of lost profits-are steadily giving way to more indeterminate standards shaped by more differentiated institutions.. and foreseeable economic' loss. Superior Court. 1984). which is administered by courts rather than discretionary agencies and which has traditionally been one of the simpler legal specialties.. 116-18 (N. 1976).. e. Petersen v. REv.2d 217.2d 334 (Cal. 163. 1983) (where threat was against mother. in which the California Supreme Court. Tarasoff. theory of recovery also included malpractice. 1984 WiS. 1968). Its relatively fixed rules-for example. in favor of a general reasonableness standard applicable even to trespassers. supra note 20. 1983) (finding duty extended to motorists injured in drugrelated automobile accident caused by defendant's patient). Also. 551 P.2d 107. Hamman v. 1976) (Tarasoff II).' in which the court jettisoned the ancient categorical rules governing the duties that landowners owe to those injured on their property. Schuck.g. amended its opinion to substitute a duty to exercise reasonable care to protect them. Even bright-line rules can become indeterminate around the edges. Myth and Reality: An Empirical Study of Private Law in Action. at 379-82. Regents of Univ. Regents of Univ. Jason Johnston not only views this decision as moving from a relatively bright-line rule to a balancing nile producing uncertainty.

C. No. 97 Stat. by reducing the largely standardless discretion that sentencing judges previously enjoyed.S. 65. 42. specialized lawyering became so essential that the profession had to reorganize along new lines. Another example is the Social Security disability insurance grid system. Other instances of simplification-through-aggregation can also be cited. It elaborated new bodies of law bearing all of complexity's hallmarks. is now widespread. 39. & 1395ww) (establishing prospective payment system under which payments to hospitals are based on fixed cost estimates rather than on the actual cost of providing care).S. 98-21. See discussion and sources cited in SCHUCK. §§ 3551-3580 (1988). simple rules. 149-63 (codified at 42 U. 40. Meant to increase certainty. See Social Security Amendments of 1983. 41. from simple rules to more complex standards. which was challenged and upheld in Heckler v." This discussion of legal forms is not meant to suggest that they have moved in only one direction. not less.41 Still. 18 U. which established a generally binding grid for judges' sentencing decisions. § 601.DUKE LAW JOURNAL [Vol. To Alschuler. Consider the Federal Sentencing Guidelines. the Guidelines are characteristic of a more general style of contemporary legal thought in which numerous cases are aggregated and treated uniformly despite their real differences. 461 U.S. the dominant trend is toward legal forms that create more discretion and complexity. Some notable and interesting counter-examples exist. 42:1 bia claims. especially horizontal equity. B. Dense norms occupied much of the field of activity. . at 31. Campbell. at 904-08. 9 Albert Alschuler and others have criticized this reform as a misguided attempt to achieve some of the virtues of clear.C. 458 (1983). L. See infra text accompanying notes 119-23. Regulatory law also brought more technicality. The Subject Matter of Legal Norms The administrative state grew out of the effort to regulate areas of activity previously governed primarily by informal or contractual norms. regulation actually reduced it on balance because of its 38. Alschuler. supra note 21. New institutions-highly differentiated agencies and the diverse structures to which they give rise 42-. and other areas previously regulated at lower levels of intensity or lower levels of government. traditionally dismissed as too speculative. Pub.proliferated. supra note 19. One important example is hospital reimbursement under Medicare according to diagnostically related groups.

Carr3 and the Voting Rights Act of 1965. 1780-1860. 44.B. 112 S.46 C.W. 1715-16 (1975). 45. Pub. and the allocation of legislative business. and egregious forms of intentional racial discrimination. Simpson. see Presley v. 820 (1992). Employment and employee benefits law. The Goals of Legal Norms Before the age of statutes. For a fundamental dispute over that degree (among other disputes). L. 186 (1962). Two examples will suffice to illustrate the point. fraud. LANGBEIN & BRUCE A. common law adjudication. the policy ends invoked by American lawmakers were generally modest. 79 Stat. See generally JOHN H." Bewilderingly complex law penetrates even private law fields." the federal law of voting rights was relatively simple. 47. Here. and an avowedly policymaking judiciary. of course. §§ 1971.S.1 (1982). The Horwitz Thesis and the History of . 1973-1973p). PENSION AND EMPLOYEE BENEFIT LAW 579-614 (1990). HORwrrz. No. 89-110. detailed agency regulation. this body of law also involves the detailed federal regulation and review of thousands of state and local election laws involving structures of representation. 88 HARV. Ct. Etowah County Comm'n. adding a new source of indeterminacy." while common law courts usually hewed to a formalistic conception of judging that emphasized (at least at the level of judicial rhetoric) a fidelity to stare decisis and legislative intent. Today. this case also exemplifies how institutional differentiation can increase complexity. rather than articulated an instrumentalist stance. concerned largely with reapportionment.1992] LEGAL COMPLEXITY greater ambition and its preservation of much of the common law regime. Before Baker v. voting rules. 369 U. is now a dense mixture of private contracting. See GUIDO CALABRESI. REV. however. 46. In addition to showing the expansion of subject matter. compare MORTON J. A COMMON LAW FOR THE AGE OF STATUTES 1 & n.49 43. and informal control through legislative oversight. For the most recent example in the Supreme Court. Stewart. It is a matter of degree. THE TRANSFORMATION OF AMERICAN LAW. The Reformation of American Administrative Law. often alluded to the law's purposes at some level of generality. the Court declined to grant its customary deference to the Justice Department's construction of the Voting Rights Act. annexations. WOLK.S. statutory norms.C. at 16-30 (1977) with A. 49. 1669. 1671-81. 437 (codified at 42 U. Legislatures' principal goals were to facilitate private ordering and to establish public policies to buttress that system. L. municipal finance. See Richard B. Judges. 48. for example. the administrative state.

the courts pursued cost-internalizing and loss-spreading goals. McDonnell Douglas Corp. 52.1717. with Horwitz's interpretation of opinions) and Gary T. See Wards Cove Packing Co. Abram Chayes. L. 642. Tort Law and the Economy in Nineteenth-Century America: A Reinterpretation. 54. 90 YALE L. 50. 105 Stat. "inventing" the doctrine of enterprise liability in the process. 93 HARV. 463 (1985). This transformation of public and private law invited judges to abandon even the pretense of formalism. 943 (1987).g. Contracts. Leaving the familiar. The Supreme Court's Due Process Calculus for Administrative Adjudication in Mathews v. Cmr. financial security for the elderly. Atonio. 533 (1979) (disagreeing. See. the more complex character of their new course quickly became clear. secure terrain of formalism for the more exhilarating but perilous realm of policymaking. The Invention of Enterprise Liability: A Critical History of the Intellectual Foundations of Modem Tort Law. 51. e.S. Eldridge: Three Factors in Search of a Theory of Value. income redistribution..54 and broad but incomplete. e. Jerry L. 792. inter alia. Stewart. Green. 648. 1721 (1981) (opinions reveal "rudimentary level" of instrumentalism. governmental reform. Once judges adopted an explicitly policy-oriented style.DUKE LAW JOURNAL [Vol.S. consistent with various purposes). . a style of thought. This multiplicity of goals necessitated new legal doctrines capable of encompassing and harmonizing them. of Community Affairs v. included health and safety. REV. See BRUCE A. 1 (1979). 1 they openly embraced a conception of role. equal opportunity. v. 450 U.J. T. development of infrastructure. Alexander Aleinikoff. L. Priest. George L. REV. RECONSTRUCTING AMERICAN LAW (1984). Fiss. supra note 48. 28 (1976). LEGAL STUD. ACKERMAN. L.' Spurred on by many legal scholars. for example. 248 (1981) (refining burden-shifting formula for all Title VII cases). In tort law. 2 The same imperatives led courts in both private and public law to adopt certain common decision methodologies that vastly complexify the system.g. REV. New legislative goals. 89 HARV. ConstitutionalLaw in the Age of Balancing. Pub. 658 (1989) (altering burden-shifting formula to increase plaintiff's burden) (effectively overruled by the Civil Rights Act of 1991. 490 U. CH. 411 U. No. 96 YALE L. v. 14 J. Owen M.S. 1281 (1976). 461. Schwartz. stabilization of markets. 802-06 (1973) (establishing . Mashaw.5 3 sequential burden-shifting. 44 U. and a rhetoric of decision that legitimated judicial policymaking in a variety of areas. Important examples include interest-balancing. Burdine. at 1678-81. largely implemented by administrative agencies. and the change it wrought accelerated in the decades that followed. and many more. a larger number of decision-relevant goals came into view.J. The Forms of Justice. 46 U. Texas Dep't. 1071). 102-166. See. 53. L. L. REV.. 42:1 The New Deal changed all that. The Role of the Judge in Public Law Litigation.

112 S. LAW AND LITERATURE: A MISUNDERSTOOD RELATION ..2d 145 (2d Cir. denied. Schuck. see Peter H. and inevitably conflicting. cert. 841. 29 C. 229. cert. affd.C. On deference to agencies.2d 37.1000 (1991) (imposing limits on 428 workplace hazards. These institutions include legislatures. Law and Administration After Chevron. and private organizations. 969 (1992).2d 924 (Cal.) (allowing liability of DES manufacturers to be based on market share rather than on specific evidence of causation of harm).S. D.. 1984) (adopting market share alternate liability theory). 452-55 (1987) (Scalia. denied. 831-32 (1992). 847-48 (1992). Litig. and other policy goals. These deference doctrines often operate differently than their authors intend. 1987). 1984) (considering market share when apportioning liability to each defendant). NLRB. The Interpretation of Legal Norms The interpretation of legal texts is an ancient problem. On deference to legislatures.. Martin v. Abbott Labs. 1004 (1988).5 By requiring judges to balance numerous diverse. administrative agencies. H.J.S.g. Ct. INS v. see. e. and the recent Supreme Court decisions in Presley v. compensation. denied. see Consumer Product Safety Act. juries. This approach (sequential burden-shifting) is designed to elicit information for the court in service of deterrence. For whatever reason-the influence of post-modernist literary theories on legal thought. Inc. 912 (1980).. In re "Agent Orange" Prod. 480 U. 492 U. these doctrines are bound to be both technical and indeterminate. and Janet Gilboy.D.. concurring in the judgment) (elaborating on the "plain meaning" approach). 53 (Wis.. Liab. in NEW DIRECTIONS IN LIABILITY LAW 4. J.S. Still. 449 U. Institutional differentiation only magnifies this complexity. Sunstein. Northwestern Bell Tel. 484 U. Occupational Safety and Health Administration Standards for Air Contaminants. On deference to juries. 826 (1984). Thomas W. § 1910. 820. v. today's Babel-like cacophony of legal hermeneutics is unprecedented.. On deference to private organizations. 251-56 (1989) (Scalia J. Sindell v.R. 421. and Lechmere. 1984) (using burden-shifting to solve indeterminate defendant problem). REV. L. policy goals. 10-12 (Walter Olson ed. and basing limits on consensus standards previously developed by private industry associations).. e. for the doctrines must be applied by agencies to whom the courts accord ill-defined deference and with whom they share adjudicatory power. Eli Lilly Co. The New Judicial Ideology of Tort Law. Judicial Deference to Executive Precedent.F. Ct. Inc. 56. Cardoza-Fonseca. 55. Supp. Abbott Labs. on file with author). 101 YALE L.. 15 U.Y. 342 N.. Etowah County Comm'n.W. 2071 (1990). Cass R. 818 F.1992] LEGAL COMPLEXITY spasmodic deference to other institutions. 832 (E. See RICHARD POSNER. Collins v.N. Co. Government Use of Private Resources in Law Enforcement (1992) (unpublished manuscript. §§ 2056(b). as the Talmud vividly demonstrates. 607 P. 112 S.56 formula for shifting burden of proof between parties in Title VII race discrimination cases). Merrill.S. 740.2d 368. 689 P. 597 F. 90 COLUM. 381 (Wash. 2058(b) (authorizing Consumer Product Safety Commission to rely on privately developed safety standards and to issue proposed rules based on such standards).. 1988).S. see. concurring in the judgment) (arguing that courts sh6uld adopt a "plain meaning" approach to statutory interpretation).J. 469 U. v.g.S. cert. See infra Section I(D).

60. at 980-85.S.. Eskridge. supra note 57. Two decisions in the Court's most recent Term indicate that this failure continues. and an enforcement apparatus which have (1988). v. supra note 55. 7 No clear winner is likely to emerge anytime soon. 467 U.C. 37 UCLA L. the affirmance rate in other circuits rose. the doctrine's growing complexity has already undermined that purpose. and if they find that it has not. 42:1 the related emphasis on interpreting norms contextually.. Constitutionalism. Whereas the D. 112 S. Chevron U. 61. a court system. the triumph of realism and instrumentalism over formalism in law. Circuit overturned agency interpretations actually rose in the years immediately following the decision. must then ask whether the agency's interpretation of the statute is reasonable. 62. the rate at which the D. Mashaw. one of the three principal pillars of the "new textualism. 841 (1992) (reversing administrative order for employer to cease barring union organizers from premises). 59. the technicality and indeterminacy of statutory norms grows. and the Interpretation of Federal Statutes.S. 58.16 DUKE LAW JOURNAL [Vol.Ct.S. Inc.. NLRB. Textualism.Ct. 820 (1992) (declining to defer to administrative interpretation of Voting Rights Act of 1965). v. at 828-29. Jr. supra note 32. 32 WM. The Remedial Infrastructure of Legal Norms Legal norms are supported and vivified by an adversary process.' Courts must first ask whether Congress has addressed the precise question at issue. Inc. 827 (1991).A. See. Lechmere. 837 (1984). Jerry L. REV. The others are clear statement requirements and "plain meaning" interpretation. Circuit initially decreased deference afforded to such interpretations. Mashaw. The rise of a "new textualism" signals uneasiness about. at 842-43. at 1041-42.62 E. Meanwhile. this new interpretive freedom.61 and even the Supreme Court has been unable to apply its own test consistently.C.60 an approach that sought to render administrative law more simple and certain by constraining the interpretive freedom of lower courts. further magnifying the doctrine's indeterminacy. REV. But less than a decade later. Contrary to Chevron's intention. Presley v." 9 In Chevron. . 57. the Supreme Court promulgated a two-part test for judicial review of agency interpretations of statutes. William N. Etowah County Comm'n. An example of this growth is the fate of post-Chevron58 judicial review of statutory interpretations by administrative agencies. 467 U. Schuck & Elliott. 112 S. & MARY L. Inc. indeed a reaction against. e. Merrill. 621 (1990). a more general delegitimation of traditional sources of authority-a plethora of competing approaches has thrown the field open. The New Textualism.g. Natural Resources Defense Council.

and that are still remediable in contract-product liability' and liability for pure economic loss are examples. Tort law has become far more pro-plaintiff since then. and more certain compensation for workplace injuries. and administrative actions with privately enforced remedies against enforcement officials as well as polluters. Workers' Compensation and Product Liability: The Interaction of a Tort and a Non-Tort Regime.E. LEGAL STUD. is now redundant in many respects. L. 66. modem civil procedure makes this a relatively straightforward matter. 488-95 (1984) (arguing for exclusive reliance on the workers' compensation system through reinstatement of the privity limitation in tort law). 13 J. Siliciano. § 2-715 (1990) (providing that following a breach of contract the buyer may recover not only for loss of a product's. Accountants' liability is an important area of expansion.economic value but also for "injury to person or property proximately resulting from any breach of warranty. L. Negligent Accounting and the Limits of Instrumental Tort Reform. e. 174 N. 475. See U.. v. pre-dispute legal counseling and documentary evidence are less comimonly available in tort law. for example. . This redundancy can be seen in the increasing overlap between tort and contract. it combines officially enforced criminal sanctions. 1929 (1988). despite the fact that workers' compensation systems are supposed to produce simpler. 63. 825 (1989).' The result is-an immensely complex hybrid claiming system. Touche.C. See Richard A. An intricate system. but the tort system's complexity and costs have also increased markedly. 1931) with H. it is replicated at three levels of government.Y. REv.") 65. tort litigation over injuries already covered by workers' compensation systems continues. causation is often difficult to ascertain. and highly differentiated remedies. 7 The complex mixture of public and private enforcement is exemplified by our environmental protection system.1992] LEGAL COMPLEXITY grown more complex with the emergence of mass litigation. civil penalties. Although the contract remedy may require a series of linked privity-based actions. Epstein. The Legal and Insurance Dynamics of Mass Tort Litigation. Compare. tort law's standard of care and measure of damages are more open-ended. Tort adjudication tends to be more complex than contract adjudication involving similar parties and subject matter. and public and private enforcement mechanisms. Ultramares Corp. 441 (N. The remedial system. At each level.J. 86 MICH. Rosenblum. Weiler. the existence of a tort law duty may be an issue whereas the existence of a contract usually is not. a bureaucratized judiciary. See generally John A. Finally. 461 A. swifter. v. This is true for several reasons.J. 5 Similarly. 1983). Adler. Paul C.g.C. Inc. Third. 50 OHIo ST.2d 138 (N. tort and workers' compensation. First. 64. 67. 3 Yet courts increasingly recognize tort causes of action for losses that traditionally sounded only in contract. Second.

19. as when a simple rule is perceived and treated as complex by those who are subject to it. a denser. The Case for Coase and Against "Coaseanism. but only up to a point. more intricate legal system may be both inevitable and desirable. Guido Calabresi & A. POSNER. 85 HARV. these costs will be quite different depending on the audience. and more difficult to reform once established. for example. and Inalienability: One View of the Cathedral. Douglas Melamed. 611. 614-15 (1989). Liability Rules. Administrators and subjects of such law must invest more in order to learn what it means. Legal complexity and legal uncertainty are cognate concepts. Cultures and markets fragment. Ellickson. but uncertainty is actually contingent on complexity. Such law tends to be more costly and cumbersome to administer. While uncertainty (or indeterminacy) is largely a function of complexity as I have defined it. THE PROBLEMATICS OF LEGAL COMPLEXITY Should we view legal complexity as a problem? If we suppose that legal rules are largely epiphenomenal. when and how it applies. which as I have noted depends in turn upon other characteristics of the legal system. See supra page 3..J. A more complex law entails many significant transaction costs which must be accounted for. 100 YALE L. Social complexity is growing remorselessly. see supra note 69. uncertainty may have other sources as well. see Robert C. Values and technologies change. 42:1 II. The Internal Revenue Code. Property Rules. 1232-34 (1991).g. See supra text accompanying notes 12. more difficult for lawmakers to formulate and agree upon. I believe. supra note 21. 70. legal complexity magnifies transaction costs by generating uncertainty. L. As Bittker has noted in the tax context." and whether the costs of complying with it are worth incurring. e. Other costs of administering a complex legal system include those related to bargaining about and around the system's rules and litigating over them. Under these conditions." 99 YALE L." For some taxonomies. Interdependencies increase. The Pointlessness of Pareto: Carrying Coase Further. reflecting the underlying social conditions to which they relate.69 Thus. REv.J. Bureaucracies expand. See. 1089 (1972). then we might want law's complexity to keep pace with society's. See infra text accompanying notes 179-80. This is true.DUKE LAW JOURNAL [Vol. ." Although complexity does not alone generate all of the costs-some would exist even in a simple legal system with 68. The Problem of Social Cost. Guido Calabresi. 1211. Coase. is a complex legal (sub)system that contains many relatively certain rules. Many of these costs are classified as "transaction costs. See supra note 10 and accompanying text. supra note 5. The systematic positive and normative analysis of how transaction costs affect legal rules is a pillar of modem legal scholarship. 69.

at 1246-47. See Samuel R. For a striking example. L. including those associated with complexity. this gap will widen. 1992). 74. create frustrating delays. supra note 20. complexity can inhibit beneficial transactions. have received much scholarly analysis-and for good reason.74 Complexity alters our incentives and tests our capacities. impose deadweight losses. POL ANALYSIS & MoMT. 85 MIcH. technology. 734. consume the energies of talented individuals. 3958. REv. which I shall empha71. at least in the short run. see Robert A. . The American Advantage: The Value of Inefficient Litigation. 10 J. 73. straining their limits. breed new and difficult-to-resolve disputes. reveal) information. of course. Such inefficiencies. To pick an extreme example. Gross. In fields as diverse as agency regulation. they are ubiquitous and limit the system's performance. complexity's demands are bound to outstrip our capacities to manage it. 748-56 (1987). Adversarial Legalism and American Government. intelligence. and unless the forces propelling complexity can somehow be reversed. impose disproportionate burdens on the poor by raising prices and necessitating the services of lawyers and other professionals trained in the management of complexity. which are in turn shaped by our values. and even if everyone knew the rule. Complexity-induced costs can be both inefficient and unfair. legal complexity can stultify a society that often depends on vigorous action in solving problems. and torts. 369 (1991) (illustrating costs of adversarial legalism with case study of the-dredging of Oakland Harbor). may have virtues as well as vices. 72. the costs of organizing a market on which these entitlements could be traded would be exceedingly high. See Hirsch. A MODEL OF THE OPTIMAL COMPLEXITY OF RULES (National Bureau of Economic Research Working Paper No. trusts and estates. lust how costly legal complexity is depends in part upon our incentives and our capacities to comply. At some point. and discourage compliance. even if the liability rule for traffic accidents was that pedestrians always win. Like friction in mechanics. Transaction costs generally. as Samuel Gross reminds us. Louis Kaplow has developed several models of legal complexity that take into account its effects on the incentives of potential violators and enforcement agencies to acquire (and in the violators' case. and social control mechanisms. creating a gap between them. Kagan. moreover.' Promoting passivity and entrenching the status quo.1992] LEGAL COMPLEXITY simple rules--it is clearly an important source of them. See LOUIS KAPLOW. But complexity increases while the constraints on our capacities-especially social values and intelligence-are relatively fixed.' Complexity's costs. But scholars have paid less attention to two other kinds of complexity-induced costs.

K. 42:1 size here. Ehrlich & Posner. Even fervent re- 75. 76. Regulatory Flexibility Act." In this locale. . As the body of rules grows more dense. one cannot survive without a great deal of local knowledge about when the buses run. unexpected cul-de-sacs. and where it is safe to walk. 552a-559. exotic cuisine. peculiar customs. its formulation entails costs. BUCHANAN & GORDON TULLOCK. those who make legal rules become more risk averse.C. each with its local patois. Guido Calabresi's image of a legal landscape constituted by the existing rules and practicesP helps to show why. This fierce localism is a vision worthy of G. the system must develop another layer of meaning: rules about rules. and jarring encounters abound. 5 U.g. Delegitimation costs occur when rules become so complex that their intelligibility and legitimacy decline. supra note 47. sudden detours.79 A legal landscape this complex engenders several kinds of governance costs that have generally been overlooked. Governance Costs When a rule must be agreed upon by a number of rulemakers. Governance costs arise from the need to formulate and gain agreement on complex rules. §§ 551. for only the initiated can lead a newcomer through the honeycomb of enclaves. 5 U. at 267.S. and belligerent pride. 77. THE CALCULUS OF CONSENT: LOGICAL FOUNDATIONS OF CONSTITUTIONAL DEMOCRACY 106 (1962).DUKE LAW JOURNAL [Vol. 78. CHESTERTON.75 A. Concealed declivities. the more costly the agreement will be.. Because integrating new rules with existing ones becomes tricky and dangerous. THE NAPOLEON OF NOTHNG HILL (1904). CALABRESI. but it is not useful to characterize a delegitimation cost in that way. §§ 601-612 (1988).S. First. arterial congestion. 79. Experienced guides equipped with maps and special know-how are essential. it is no longer enough to know one's location and destination. Chesterton. E. supra note 20. See GILBERT K. A governance cost can readily be viewed as a kind of transaction cost.76 A legal system that is complex in any of the senses I have defined will multiply these governance costs. puzzling signs. which vendors are trustworthy. whether cabs will venture into certain neighborhoods. the legal landscape becomes more thickly populated and harder to traverse. 701-706 (1988). the more numerous those whose assents are necessary. Administrative Procedure Act. at 3-7.C. JAMEs M. People with a choice do not venture out unattended.

complex standard: Judges of radically different political and social viewpoints can only reason together on the basis of a few relatively simple prop80. Eustice. another governance cost.45 TAX L. The proliferation of policymaking institutions multiplies the sources of innovation. I discuss the phenomenon in PETER H. see. ECON. Lon L. any change will have ripple effects.19921 LEGAL COMPLEXITY formers hesitate to alter a landscape that is so hard to read. especially if it is itself a collegial or otherwise differentiated body. Compare Jerry L. The Forms and Limits of Adjudication. Tax Complexity and the Tax Practitioner. 82. information. rulemakers are more likely to adhere to even an unsatisfactory status quo. REV. usually agencies and courts. SUING GOVERNMENT: CITIZEN REMEDIES FOR OFFICIAL WRONGS 70-71 (1983). Institutional differentiation spawns legal indeterminacy.. they know that in a more polycentric legal world. When Does Worker Ownership Work? ESOPs. This is a logical implication from official risk aversion. that exists today in so many policy fields. Codetermination. 99 YALE L. & ORGANIZATION 81.J. . 83. On the other hand. REV. 7. 1979) (criticizing failure of Congress to make policy decisions). Judge Patricia Wald of the District of Columbia Circuit recently underscored this point when she defended the concededly problematic Chevron test to those who would prefer a more refined. These standards do not really reduce conflict but simply use delegated authority to move the policy conflict from one rulemaking locus to others.' When the risks of error are magnified.L. 11 (1989) (leaving unanswered tax policy questions to courts shifts complexity problem without eliminating it). The only way to reach agreement among institutions may be to adopt open-ended. multi-factored or otherwise indeterminate legal standards. see infra text accoml~anying notes 107-11. Law Firms. With respect to public actors. LOWI. 353. swiftly. and unpredictably throughout the system's web. Henry Hansmann. ramifying widely. Mashaw. THE END OF LIBERALISM: THE SECOND REPUBLIC OF THE UNITED STATES (2d ed. 1749. this diversity also encourages conflict and raises decision costs. 97 (1985) (arguing that "[e]limination of discretion at one choice point merely causes the discretion that had been exercised there to migrate elsewhere in the system") with THEODORE J." This increased risk aversion surely contributes to the bleak policymaking deadlock. 1 J. eg. L. 81. With respect to private actors. see also James S. 395-404 (1978) (describing concept of polycentricity). 1780-83 (1990). Fuller. 92 HARV. and Economic Democracy. SCHUCK. and legitimacy-precious resources in any social system. the sense of drift. Prodelegation: Why Administrators Should Make Political Decisions." Delegated standards thus increase the delegate's cost of decision.

Delegitimation Costs I have been discussing costs of legal complexity borne in the first instance by those who must administer the system. 94-95. in the forms described earlier: blocked transactions. of course. at 49-51 (citing JOHN FINNIS. This common sense. Again. warts and all. deadweight losses. 5 But if the complex legal landscape contains many pitfalls for the governors. common experience. NATURAL LAW AND NATURAL RIGHTS 100-27 (1980)). Michael J.' B. our debates would be less focused. flow through to the rest of us. nor can their applications be readily predicted. 42:1 ositions that they can all agree govern. however. especially valuable in connection with complex rules. they will often be opaque to the common mind. 9 Intelligible only to experts. REv. and our differences would more frequently prove irreconcilable. 767 (1977). This concept is elaborated in ISAIAH BERLIN. FLA. Patricia M. If the rules are technical. common sense. so that disputes can be narrowly focused on their application." a term coined by Bayless Manning.. is a public good insofar as it creates precedents that clarify and refine the law for the use of others. heightened risk aversion. Wald. OLSON. 668-69 (footnote omitted). . litigation.. 1991. supra note 20. 89.g.' When rules are indeterminate.. 86. at 94. delays. it is terra incognita for the governed. the law is likely to mystify and alienate 84. Cacace. 71 Nw. their precise meanings cannot be easily grasped. Hyperlexis: Our National Disease.J. and conflict among governing institutions. 88. this is a bootstrapped benefit.. In a society that values negative liberty"6 and personal "space. 1991 DUKE LJ. See. Litigation. To the extent that the rules are more complex than they need to be. U. and even common morality. e." the density of the legal system-the penetration of law into every corner of human life. 85. supra note 9. Confusion aid uncertainty follow. The Nation's LoveHate Relationship with Lawyers. describes "the pathological condition caused by an overactive law-making gland. or what has been called "hyperlexis"---is bound to be a source of deep resentment. B.DUKE LAW JOURNAL [Vol. a benefit that is. 767. Comment." D'Amato. of course. in FOUR ESSAYS ON LIBERTY 118 (1969).. 87. for example. L. this does not deny that some of these effects entail countervailing benefits. The "New Administrative Law"--With the Same Old Judges in It?. Two Concepts of Liberty. Oct. our work would be more complex. 647. common morality criterion is akin to what Anthony D'Amato calls "natural law" and John Finnis calls "practical reasonableness. A simple decisional framework may be essential to collegial decisionmaking in a court of widely differing views. These costs." Bayless Manning. "Hyperlexis. at 1-11. The Chevron decision. Were we to switch to a broader framework of what is good regulatory policy or which of thirty canons apply. has accomplished this .

they will come to view complexity in an altogether different. 93. 92. The fact that these characteristics of the complex legal regime seem to be objective may for a time earn it acquiescence. in 2 TAXPAYER COMPLIANCE 156. even support. Our ignorance extends even to the tax field. the public suffers excessive complexity in silence. Compliance Costs. See infra Part III. But as we shall see. responds with informal simplifications. Historical studies indicate that popular disgust with legal complexity and a quest for simplification was an important factor leading to the adoption of code pleading in this country. I suspect. where understanding this relationship is an important policy priority. distributed.9 This successful reform of the formal law-what John Langbein calls "legal implosion"-is surely exceptional.1992] LEGAL COMPLEXITY lay citizens whose intelligence it often seems designed to mock. See infra Part IV. Profound cynicism about and alienation from the legal system may result. and Tax Evasion. see Susan Long & Judyth Swingen. 178 (Jeffrey A. 1949). however. at 22.. its legitimacy-the sense of "oughtness" that the lawmakers hope will attach to it-is diminished. E.94 91. rather than randomly. of course. in DAVID D. Understanding Taxpayer Compliance: Self Interest. One cannot measure these delegitimation costs directly. An Approach to the Measurement of Tax Law Complexity. FIELD: CENTENARY ESSAYS 17. that these delegitimation costs are great. and that this distribution is purposeful rather than adventitious. Roth & John T. 94. For a discussion of some of the methodological obstacles to such studies. Jeffrey A. Alison Reppy. and tax evasion"). complexity is not neutral in its effects. AM.. 21-25 (Alison Reppy ed. Spring 1987. .' or simply fails to comply. Complexity.' a fact that is well understood by those who seek to influence the shape of the law. Taxpayers bewildered by tax law's complexity and uncertainty appear more likely to violate it.. Once citizens begin to suspect that these advantages and disadvantages are unfairly. "only recently has research been focused on the relationship between complexity.93 There is evidence.. TAX ASS'N. The literature reveals remarkably little empirical inquiry into the link between complexity and noncompliance. Joel Slemrod.g. it advantages some groups and disadvantages others. 25 (proposing six indices of complexity). compliance costs. The Field Codification Concept. less tolerant light. More commonly. J. Roth et al. Scholz eds. When this Delphic law also emerges from an institutional black box that is itself dense and difficult to comprehend. much less know the extent to which legal complexity is their cause. 1989) (reviewing methodological issues.

Dave Barry. e. TIMES. Apr. the temptation to use self help to fashion their own private tax shelters becomes well nigh irresistible. plus it is used to take taxpayers' money away and turn it over to a federal government that routinely spends it on activities like declaring National Deviled Ham Awareness Month. Mar. Humorist Dave Barry imagines that taxpayer noncompliance has something to do with the fact that the tax code is larger than the average taxpayer's home and nobody in the entire world really understands it because it was apparently written by hostile mutant non-English-speaking lawyers from space. 98."95 There is ample evidence of delegitimation costs in fields other than tax.. 129 ("Of the seven guilt neutralization mechanisms investigated by Thurman . in 1 TAXPAYER COMPLIANCE. "Erosion of taxpayer morale. TIMES. legal complexity has helped to produce an immense illegal sector. 4-5. the justification. 1992. so critical to the tax law's integrity. Social Commitment. at B1 (candidates subject to summary removal from ballot for spelling errors on ballot. 97. off the primary ballot. See Michael Specter. 42:1 Voiuntary compliance. 11. Schuck & Robert E. Nah. at 18. INT'L HERALD TRIB. 1992.. N. a genuine presidential contender. and confusing signals to manufacturers that it had little effect on the product design decisions it was supposed to influ- ence. complexity has helped to delay. Once Again. plus it is revised about every 25 minutes. etc. 96. WIrE. "is not an unimportant side effect of all this confusion-if taxpayers think the whole system is unknowable. GEORGE EADS & PETER REUTER. 6 New York's hypertechnical signature and petition require- ments nearly kept Paul Tsongas. only by invoking another technicalityY Recently. HERNANDO DE SOTO. random. a result that was averted." In the United States. and Other Influences. ironically. at A18 (lawsuit challenging Tsongas dismissed on technicality). It's probably some kind of chemical disorder in the taxpayers' brains." James Eustice observes.). 28. supra note 93. found that the law emitted such noisy..' was the most widely accepted and most consistently related to compliance. N. depends upon its intelligibility.") 95. wrong type of clips used to bind pages. Litan. at 19 (but citing no empirical support for proposition). THE OTHER PATH: THE INVISIBLE REVOLUTION IN THE THIRD WORLD (1989). Now It's a Technicality That Keeps Tsongas and Brown on Ballot. THE POLITICS AND DEVELOPMENT OF THE FEDERAL INCOME TAX 62 (1985) (arguing that complexity "makes the task of enforcement significantly more difficult. Schedule 936850345-D. for example. see also JOHN F.Y. Feb. supra note 83. incorrect ordering of witness sheets. Regulatory Reform in .Y. See. that couldn't be it.g. at 71. in some Latin American societies.. 1992. A Rand study of corporate responses to modern product liability law.DUKE LAW JOURNAL [Vol. and it makes it easier to plead ignorance or error when the intent was truly fraud"). 'It is okay to claim an undeserved tax deduction when you are not really sure what the rule is. DESIGNING SAFER PRODUCrS: CORPORATE RESPONSES TO PRODUCT LIABILITY LAW AND REGULATION 27-29 (1985). Michael Specter. Peter H. Eustice. New York's Arcane Election Laws Shape Race.

Law Abounding: Legalization Around the North Atlantic. The two are closely related. e. In a real sense. legal complexity is simply not preordained by social complexity. 1982). Lindblom. Robert Pear. I underscore the disadvantages of complexity here only because the rulemaking process systematically obscures them. See. legal commentary. and that this strategy is both rational and typical.. however. Higher levels of education and professionalism have increased the public's taste and tolerance for complexity in all social institutions. legal techniques. Complexity. might be even worse. Why and in what sense do we choose it? One can argue that the the Third World. 99. at Al (noting delays in legislation on the homeless.S. 55 MOD.. The social conditions that the law seeks to regulate have grown more difficult to understand and control. 51 (1986). these weighty objections to complexity would not necessarily be decisive in any particular situation. we choose it. Research in experimental psychology and organizational behavior indicates that they radically simplify their complex decision environments in order to make them manageable. response to social complexity. ON REG. Mar. e. I discuss some of these examples in Part IV. THE POLITICAL ECONOMY OF LEGAL COMPLEXITY Legal complexity clearly has many causes. at 80.).. litigation. the implementation of many important regulatory statutes. A simpler regime. Social actors manifestly understand this. that more social complexity leads inexorably to more legal complexity. 4 YALE J. U. but one does not always follow from the other. after all. III. 1 (1992)..99 Again.Y. N. of course. is subject to the law of diminishing returns. problems deemed fit for legal regulation-have expanded enormously. however. or at the least a plausible.g. SIMON. See.g. Marc Galanter. supra note 5.g. etc. eds. 31. 100. lawyers. Laws Delayed by Complex Rules and Partisanship. like most other things. See. supra note 5. . at 81-83. The economy has grown larger and more interdependent. including the legal system. 1991. waste storage. JUDGMENT UNDER UNCERTAINTY: HEURISTICS AND BIASES (Daniel Kahneman et al. REV. TIMES.'0 ' Instances abound in which legal simplification seems the best.1992] LEGAL COMPLEXITY and in some cases defeat. All of the components of the legal world-law. Beyond a certain point. L. 101." One must not assume. e. then. The Case of Peru. air pollution.

and that this demand may sometimes carry the day.g. L. e. 6 J. and Critical Stories of Legal Development. By rooting their self-interest in rather conventional professional. ECON. legal complexity is not so much the goal of any individual actor's choice as it is the indirect consequence of a multitude of decisions made on other grounds. nontautological sense. I noted earlier that the benefits and costs of complexity are not evenly distributed and that this uneven distribution is hardly an accident. It is to claim. 167 (1988). that the burden of political inertia usually ensures legal complexity's triumph. I hope to minimize the force of this objection. rather. & ORGANIZATION 267.. The Theory of Public Choice. REV. See. 3 The beneficiaries of complexity are many. bureaucrats.L. and legal scholars-generally benefit from legal complexity while bearing few of its costs. In this story. Symposium. I wish to offer a different account-one of internal rationality. . Mashaw. Positive. Explaining Administrative Process: Normative.DUKE LAW JOURNAL [Vol. In this "external rationality" account. 74 VA. See infra text accompanying notes 106-17. 102. lawyers. 269-72 (1990). This external rationality story surely explains a good deal of legal complexity. directed at some versions of "public choice" theory. that self-interest can be defined self-referentially and nonfalsifiably. such as a desire to address a new condition or to accommodate a hard case. groups that complexity insulates from public accountability. This is not to deny that an aroused public opinion can occasionally be mobilized to demand greater legal simplicity. rationalizers. and groups that supply the services that complexity renders valuable. I recognize the often valid criticism. 103.l"c On balance. as the code pleading movement did in the nineteenth century. even if they cannot always precisely gauge the relevant magnitudes and distributions. These decisions then combine to yield a complexity that survives because individual actors lack the incentive to change it. careerist. Although I mean to use self-interest here in a fairly strong. They include groups that are relatively well equipped to cope with complexity and for whom complexity can create a competitive advantage outweighing its costs. they prefer a complex system and hence choose rules and institutions that tend to support and perpetuate it. the main producers. Political actors understand this fact and seek to exploit it. and administrators of law-legislators and their staff. I shall specifically discuss these groups in a moment. litigants. 42:1 social choice for complexity is not a conscious one. and economic goals. judges. Jerry L.

Given their obvious economic stakes in underlying substantive issues (say. are even more numerous. Legislators might prefer legal complexity for four self-interested. political opponents can easily stigmatize their complaints about complexity as self-serving and pretextuous. the beneficiaries of complexity can drape themselves in lofty public interest goals. conferring divisible policy benefits on one or a few groups without having to confer them on all. (3) enhance their power over bureaucrats. such as securing the individual's right to a day in court. one can readily predict that many worthy simplification reforms will die on the vine and that others will succeed only if they happen to coincide with the interests of politically powerful groups or with the government's own fiscal stakes. narrowly focused regulated interests. Complex statutes enable legislators to create private goods. See generally STEvEN KELmAN. When the interests of law producers are added to the equation. Legislators who wish to target a 104. and the like-are left to demand simplification. For two such examples see supra note 41. (2) confer divisible non-policy benefits. public interest arguments for policy change. Thus. preventing the shrewd from circumventing the law. Free rider and other obstacles to political organizing will likely dissipate most of their energies and pressures. 5 Legislators and Their Staff. Only those cost bearers with the most at stake-insurers and other institutional litigators. Complexity can help them to (1) confer divisible policy benefits on constituents. the political economy of legal complexity contains some structural elements favoring its survival and flourishing.19921 LEGAL COMPLEXITY Those who ultimately bear complexity's costs. of course.1° In contrast to the cost bearers. yet they labor under a serious political disadvantage. mass tort claims or natural gas prices). . electorally related reasons. and heading off problems before they arise. and (4) ease the legislature's collective action problem. These cost bearers have an incentive to press politicians and other law producers to simplify the rules of the game-but they constitute an exceedingly diffuse group. MAKING PUBLIC POLICY: A HOPEFUL VIEW OF AMERICAN GOVERNMENT 248-70 (1987) (examining the role of public spirit in the political process). Complaints about complexity will seem especially disingenuous in a political ethos like ours that increasingly demands disinterested. 105.

On the Accelerating Rate and Decreasing Durability of Tax Reform. 139 U.Y In addition to (or instead of) targeting policies at particular constituencies.C. Legislators can. FIORINA. In these ways. REV. For example. a prominent political scientist has argued that legislators' create complex new programs and bureaucracies in part so that they can intervene selectively 1 later on when problems materialize. L. 42:1 statutory . less discriminating statutes do not usually permit.. achieving a level of control that simpler. 1991. oppose a new benefits policy yet intervene on behalf of constituents later on when the agency denies their claims. Organizing for Visas: Irish.Y. This "casework" with the relevant bureaucracy permits legislators to confer highly visible." Complex laws also enable legislators to exercise influence over the personnel."° Recent immigration legislation accomplished a similar sleight-ofhand. at 31. Beyond Public Choice and Public Interest: A Study of the Legislative Process as Illustrated by Tax Legislation in the 1980s. e. Immigration Act of 1990. II 1990) (providing that no less than 40% of the visas during a three-year transitional period must go to "the foreign state the natives of which received the greatest number of visas issued under section 314" of the 1986 law). 71 MINN. N.g. 12. e. REV. MORRIS P. 8 U. and establish new institutional power centers. complex laws also create opportunities for legislators to engage in targeted intervention at the law application stage.DUKE LAW JOURNAL [Vol. they can hope to calibrate their political support. See. Matthew D. 1 (1990) (criticizing public choice model and emphasizing prestige and symbolism explanations of recent legislation). classify groups."° Such influence is politically valuable 106. policies. That "foreign state" was Ireland. 1989). Oct. Administrative Procedures as Instruments . Indeed. 109. CONGRESS: KEYSTONE OF THE WASHINGTON ESTABLISHMENT 44-47 (2d ed.McChesney. 108. Richard L. See. McCubbins et al. TIMES.. Daniel Shaviro. 913 (1987) (public choice model attributing recent increase in the rate of tax reform to shorter-term "contracts" between legislators and private interests). Lottery and Luck. politically riskier policy issues. selective benefits on favored constituents without limiting their own freedom of action on larger. See Donatella Lorch. they can multiply and refine policy categories. 33. for example. Doernberg & Fred S.policy to favor (or disfavor) particular interests can employ a number of complexifying techniques. § 132(c). and budgets of the agencies and the programs that they create.S.g. Without quite naming the Irish. L. Tax legislation is notorious for this.. 107. Complexity can also help to obfuscate certain political goals (such as wealth redistribution) that would invite opposition to policies that seemed to implicate them. PA. the statute created a special new preference for them and concealed it in a remarkably complex and opaque formula. § 1153 note (Supp.

"3 Although their incentives largely parallel those of their elected employers. and also explains much of 1 ° their post-enactment.569 in 1989. minimize rulemaking costs. and enhances their personal satisfaction and career prospects.. ORNSTEIN ET AL. coupled with other changes in legislative operations at all levels of government."' Ironically. staff members also have strong personal. L. and is also essential in assuring bureaucratic compliance. By placing complex procedural constraints upon an agency's decisionmaking process. The potential for this influence provides a powerful motive for legislators to enact complex statutes. Complex legislation enables staff members to enlarge and exhibit their professional and technical expertise. 114. UNELECTED REPRESENTATIVES (1980). VITAL STATISTICS ON CONGRESS. 431 (1989). magnifies their influence over policy and administration. 112. Matthew D. see FIORINA. legal complexity can help to solve politicians' problems by creating new problems for their constituents. DIRTY AIR (1981). See supra text accompanying notes 80-83. at 172-78 (discussing workload which in part re- .055 in 1967 to 7. ambiguous provisions. CLEAN COAL. politicians are able to respond to potential deviations from the status quo prior to the implementation of a change in policy.. In recent decades.19921 LEGAL COMPLEXITY in itself. and careerist reasons to draft more complex bills.S. has led to staff dominance of much of the legislative process and output. 111. REV. Politics and Policy: Administrative Arrangements and the Political Control of Agencies. For a discussion of the concerns raised by the explosion in Congressional staffing. professional. Legislators who wish to reduce conflict.L. the number serving on members' staffs in the U. 4 of Political Control.749 to 3. and delegation of rulemaking authority to other more specialized institutions. then.837 in the same period. MALBIN. NORMAN J. 2 This growth. 1991-1992. activity. at 126 (1992). see BRUCE A. For some revealing case studies. their number has increased dramatically. supra note 108. The number of aides serving on members' personal staffs in the U. A final consideration concerns legislators' staff aides. Structure and Process. Examples include logrolls.S. 113. 110. Legal complexity can help legislators avoid collective action problems that a simpler law might well exacerbate. ACKERMAN & WILLIAM T. MICHAEL J. See MALBIN. 3 J. McCubbins et al. and thereby retain effective control over agency decisions. deepens legislators' dependency on their advice. House of Representatives rose from 4. HASSLER. and reach consensus in the face of their strong incentives to behave opportunistically often resort to complex legislative forms. & ORGANIZATION 243 (1987). Senate rose from 1. ECON. omnibus bills. at 118-21. 75 VA. supra note 113.

JR. 97-261.g. No. For somewhat different reasons. agencies constrain the ability of regulatees to either elude agency jurisdiction or circumvent agency standards. 96 Stat. Pub. and other transitional and compensatory mechanisms7 in order to move from a complex regime to a simpler 11 one. 1705 (codified in scattered sections of 49 U. Motor Carrier Act of 1980.). Johnston points out that even when agencies issue relatively precise rules. fairness. 16 This outcome is not as paradoxical or unusual as it might seem. has actually increased the law's complexity in important respects."' By filling most of the relevant policy space with rules. The Tax Reform Act of 1986. EUGENE STEUERLE. 117. they nevertheless possess a virtue that often proves decisive: they preserve agencies' future freedom of action in policy environments dominated by uncertainty and the need for flexibility.). 132 (codified in scattered sections of 12 U. No. No. 96 Stat. Dense rules are designed to foreclose such tactics. Gan-St. 97-320. No.C. 118. at 958-59. detailed transitional rules necessary to implement the new policy against real estate tax sheltering. 115. 94 Stat. Politics. L. This is especially true of the intricate. Pub. Although costlier for agencies to apply and for regulatees to apprehend. 100 Stat. e. supra note 106. THE TAX DECADE: How TAXES CAME TO DOMINATE THE PuBLic AGENDA 116-17 (1992). 94 Stat. added to the complexity and uncertainty faced .S.). Airline Deregulation Act of 1978.C. L.C. 793 (codified in scattered sections of 49 U. See Depository Institutions Deregulation and Monetary Control Act of 1980. 92 Stat. No. Railroad Revitalization and Regulatory Reform Act of 1976.. L. BUREAUCRACY AND REPRESENTATIVE GOVERNMENT (1971). Germain Depository Institutions Act of 1982. L. 94-210. Pub. 1102 (codified in scattered sections of 49 U. 116.). if anything.S.S. side payments. 119. L. Bureaucrats..S.DUKE LAW JOURNAL (Vol. Bus Regulatory Reform Act of 1982. NISKANEN.115 which was sold to the public as a simplification measure. See Doemberg & McChesney. See. 31 (codified as amended in scattered sections of 45.C. Pub. 1469 (codified in scattered sections of 12 U. Their success in doing so is of course the subject of innumerable jeremiads against regulation. Pub.).C. WILLAM A.S. Pub. No. the courts' refusal to recognize compliance with such rules as a defense to tort liability has. L.S. Pub.). and economic efficiency often require policymakers to adopt grandfather rules.C. Indeterminate rules also serve important bureaucratic interests. 95-504. 2085 (codified as amended in scattered sections of 26 U. No. agency officials possess strong incentives to elaborate legal networks that meet all of my complexity criteria.S. 99-514. 90 Stat. 49 U. Agencies may also want the rules to leave some flexibility for the regulated even at the price of some uncertainty. L. 96-221.). 9 flects complexity). See C. 96-296. 42:1 Perhaps the most revealing evidence of the legislative impulse toward complexity is its ability to prevail even when a constituency for simplification exists and public pressures for reform are strong.C. subsidies.

See supra note 20 and accompanying text. Johnston." This cohort does not ordinarily dominate a company's decisionmaking process. See MASHAW & HAEPST. McGarity. and Judges. and managing it. Autumn 1991. LAW & CONTEMP. a specialized bar."' These entities then take on lives of their own. which interact with agency activities to promote additional complexity.. and related requirements). at 57 (describing the structural and decisionmaking apparatus of the Environmental Protection Agency). but it does act as a kind of pro-complexity Trojan Horse inside the corporate structure. cost-conscious resistance to complexity. Lawyers. technical rules promote agency autonomy. 121. and many others. Their endemic need to mobilize outside information and political support prompts them to proliferate new entities: coordinating mechanisms. 123."H Indeed. of all the causes of legal by regulatees. 120. See Thomas 0. which coordinate compliance with health. complying with. PROBS. administrative review panels. agencies are not merely instances of institutional differentiation. In addition. Finally. WILSON. As noted earlier. the nature of their tasks ordinarily ensures that their rules will usually be intelligible only to cognoscenti in the field. weakening somewhat management's more reflexive. The Internal Structure of EPA Rulemaking. at 117 (auto manufacturers devote substantial resources to regulatory affairs offices. they are also causes of it. technical support groups. there is a set of corporate officials whose jobs and occupational advancement depend upon understanding. 2 ' Being more opaque to the generalist institutions like Congress and the media that seek to influence it. BUREAUCRACY: WHAT GOVERNMENT AGENCIES Do AND WHY THEY Do IT 244-48 (1989) (technical task structures make agencies more independent of congressional control). at 366. such rules make agencies more difficult to control and help obscure their pursuit of controversial policies. supra note 20. Litigants.. the question of how the litigation process shapes legal rules has received considerable scholarly attention. For every new regulatory program or legal complexity.19921 LEGAL COMPLEXITY Agencies need no special incentives to elaborate technical rules. advisory committees. The same is true of complex procedures that make agency decisions more costly to challenge. 122. intergovernmental bodies. See JAMES Q. supra note 13. These agency interests in complexity are mirrored in the interests of some of those in the private bureaucracies that agencies regulate. a specialized press. . safety.

show that disputes turning on determinate rules will be settled rather than litigated (regardless of their content). Some scholars. at 311. he says. 128. Johnston." Michelle White argues that tax lawyers. at 371. supra note 20. and that this trend is a one-way ratchet: courts. supra note 20. but for different reasons. 39 UCLA L. D'Amato. LEGAL STUD. supra note 20. Lawyers. enhance the settlement value of plaintiffs' claims. for example. Henderson. and a rule that disadvantages tends to invite challenge. who seek to maximize demand for their services by promoting more complex rules that invite expert interpretation and manipulation. can only realize this goal through their clients' litigation. 78 AM. 4 Others. Jr. like Jason Johnston and Carol Rose. Carol Rose.125. They emphasize that such rules. 1 (1984). and by legislators and transactors who must then try to restore a regime of crystals in the interests of efficiency. pt. focusing on property law. Rose."z Making a related point. Although he does not go on to say that lawyers never subordinate the clients' interests to their own. ECON. George L. at 23-24. . Priest and Klein. discern an inexorable trend toward indeterminacy. Johnston. also perceives oscillation. he does argue that the lawyers might benefit more from oscillation than from a steady trend toward indeterminacy. litigation dynamics appears to be the only one that has been analyzed systematically. Jr. which assures that the rules litigants bring to court are at the indeterminate end of the spectrum. 42:1 complexity. find it hard to go back to more precise ones. but ascribes it to the incentives of lawyers.. have vested inter- 124. Epstein. REV. 126. but for her it is driven by judges who muddy the crystalline rules to protect the "fools" ensnared by them. The Quiet Revolution in Products Liability: An Empirical Study of Legal Change. James A. the resulting litigation systematically favors the evolution of more indeterminate rules. 479. The Selection of Disputes for Litigation. however. Johnston criticizes Epstein's view on several grounds.questions. Henderson. 313 (1988). at 371-72. 127. The Political Economy of Product Liability Reform.DUKE LAW JOURNAL [Vol. at 603-04. 13 J. Richard A. by creating jury . Priest & Benjamin Klein. argue that incentives endogenous to the litigation process generate a pattern of perpetual oscillation between bright-line and open-ended rule forms."6 Henderson and Eisenberg also claim that indeterminate rules tend to favor plaintiffs. like tax accountants and the Internal Revenue Service. & Theodore Eisenberg. 731 (1992). 37 UCLA L. REv. having adopted indeterminate rules. 515 (1990). Anthony D'Amato argues that because a rule is easier to challenge than to defend. 2. Inside the Quiet Revolution in Products Liability. . see also Theodore Eisenberg & James A." Richard Epstein likewise sees a firm trend toward indeterminacy. REv. supra note 31.

L. at 10-11. This. REV. Malone. Why Are Taxes So Complex and Who Benefits?. Wold & John H. whereas other models ascribe greater importance to judges' own goals and their discretion to pursue them. 45 TAX L..'32 Judicial innovations tend to make the law more complex. 47 TAX NOTES 341 (1990). call attention to herself in order to attract support from special interests. however. 99 (1956) ("Thus distinction is heaped upon distinction and law becomes increasingly and unnecessarily complex. 1991). 413 U. judges can innovate secure in the knowledge that it will be up to anonymous jurors to determine the concrete meaning and consequences of the change.' although there are certainly exceptions. although each group favors a different level of complexity. see Boris I. Gonzales. 435 U. at 11.S. James S. Schuck. REv. Chadha. which is somewhat less determinate and could send more cases to the jury. See.1992] LEGAL COMPLEXITY ests in a complex tax law. should be seen as the extreme case in which complexity's limits have manifestly been reached. Miller v. Baker v. 15 (1973) (moving to less technical rule).g.g. With a jury. See. 1 (1989). For a Bittkeresque parody of the . 29' Boris Bittker suggests that since tax experts' analytical and critical skills account for much of the law's complexity.2d 872 (Cal. they cannot be expected to make it simpler and more enforceable. Inc. 186 (1962) (moving to more bright-line rule). INS v. 133."' I endorse the latter. Bittker. See. Henderson & Eisenberg. supra note 15. White." ° Some litigation models view judges as merely ratifying the moves of the parties and their lawyers. Eustice. e. 60.' Some judicial innovations are dictated by traditional criteria. supra note 125 (supporting the former model). and in some states reelection can be a very risky and expensive business. 919 (1983) (moving to institutional simplification). or distinguishing precedent. 519 (1978) (moving to less density). See Priest & Klein. but many others reflect different motives. refining. 132.S. Culver. inde- 129. 819 P. Ruminations on Cause-in-Fact. California. 462 U. 135.S. John T. 9 STAN. Wex S. e. e.. A recent example of this is the en banc decision of the California Supreme Court in Mitchell v. The Defeat of . Vermont Yankee Nuclear Power Corp. supra note 127 (supporting the latter model).. in which the court replaced the "but for" test for cause-in-fact in negligence actions with the "substantial factor" test.S.g. The judge may wish to reduce her boredom.notion that the Code's complexity reflects nothing more than profit-maximizing by tax professionals."' or burnish her reputation for creativity. and I therefore discuss it in Part IV. 130.. supra note 55. 131. Bittker. Carr. Natural Resources Defense Council. v.") 134. 369 U. Most American judges are elected. having suggested elsewhere that the jury reduces the cost to judges of innovating--of rejecting. Perhaps the most notable simplifying innovation is the current effort to simplify asbestos and some other mass tort litigation. Michelle J.

136 Although the interests favoring complexity are systematic and powerful.U. 70 JUDICATURE 348 (1987). Doctrinal innovations. Jan. supra note 20. 138. Texas Court Fight Puts Focus on Elected Judges. 22. It is only that these interests tend to predominate. see Michelle J. 42:1 pendence. 48 (1988). some of them will be disadvantaged by complex substantive rules-although it must be added that most clients play little or no role in the formulation of legal arguments. who must bear most of the costs of litigating under complex rules and procedures. moreover. The Independence of Lawyers. may prefer simplicity out of narrow self-interest. 39 As for judges. 1. the Electorate. are also likely to enhance what D'Amato has called the "information value" of a decision. academics. REV. legal scholars are drawn to legal the California Justices: The Campaign. at B4. 1988. Legal Scholars. TIMES. Gordon. L. HOROWITZ. then. Like other denizens of the legal culture. THE JUROCRACY (1977) (describing powerful role of lawyers in administering government programs). and other court-watchers. 137. 139. . 40. D'Amato. 68 B. White. Because litigants bear many of the costs of a complex adjudication process. and that this fact is an important reason for the growth in legal complexity. Robert Gordon. an important consideration for overwhelmed courts.' Even lawyers who would benefit financially from a more complex law may nevertheless support simplification reforms for public interest reasons. at 19-22 (describing information value as a measure of the divergence of an actual message or result from an anticipated message or result). and professional skill with lawyers. journalists. argues that this form of professional public spiritedness has declined over time. 37 Contingency fee lawyers. My claim. N. while noting that "historically lawyers have sacrificed income repeatedly" in order to advance other goals. they are not monolithic. see also DONALD L. Robert W. See generally DOUGLAS E. who can draw upon their own in-house legal expertise and who also are repeat players. 1989) (unpublished manuscript on file with author). Peter Applebome. For an economic model that predicts that contingency fee lawyers are the group tending to prefer the simplest law. and judges do have countervailing interests in promoting simpler law. at least some of the time.DUKE LAW JOURNAL [Vol.Y. This is less true of corporate clients. and the Issue of Judicial Accountability. LAWYER AND CLIENT: VHO'S IN CHARGE? (1977) (describing relationship between lawyers and their clients in personal injury cases). Some litigants. 136. is not that the interests of litigants. but for altogether different reasons. lawyers. and judges in legal complexity are invariably and unalterably opposed to greater simplicity in law. ROSENTHAL. lawyers. simple rules are easier to administer and generate fewer disputes. Legal Complexity (June 30.

and law and economics-occupy the academy's ideological center. if seriously pursued.19921 LEGAL COMPLEXITY complexity. and right. see William N. I called it a craft value. complexity is aesthetically pleasing. and more satisfying than simplicity generally permits. which views and seeks to understand reality as an intricate array of fields on which congeries of forces interact in hard-to-predict ways. policymakers (whose biases toward complexity are not intellectual). This tendency I believe reflects more the logic of the scholars' theories than any 140. always great. . For a recent. left. scholarly superstars. My characterizations of these schools in this section draw heavily upon this account. We delight in stories that are richer. This commitment. and distinguished university departments are made. and the subversion of traditional paradigms-is the supreme academic virtue. It also reflects a uniquely scholarly commitment. is growing apace in the law schools. 142. Schuck. Jr. 707 (1991). Earlier."' they have been that much busier seeking to elaborate novel theories and subvert traditional paradigms. This taste. Why Don't Law Professors Do More Empirical Research?. L REV. Three "schools" of legal theory have dominated recent legal scholarship.1" There are reasons to think that this distinctive scholarly commitment to complexity. 323 (1989). the stuff of which Nobel laureates. 141. more encompassing. would tend to make the law more complex. Like our competing attraction to elegantly simple theories. the elaboration and testing of novel hypotheses. sophisticated folks. Creativity-the discovery of new information. These schools-legal process/judicial policymaking. 42 No general characterization of these schools can fully capture the diversity and nuanced positions within them. LEGAL EDUC. respectively. The most important of these groups are the media (whose attitudes toward complexity are shaped by the technical constraints). is not merely aesthetic. and what used to be called the intelligentsia (when Marxism was more in fashion). distinguishes scholars from other well-educated. The New Public Law Movement: Moderation as a Postmodern Cultural Form. 89 MICH. Eskridge. perspicuous discussion of the three schools. although they do not begin to exhaust the set of approaches to legal analysis. as much as anything else. Perhaps because legal scholars are generally less interested than their natural and behavioral science colleagues in discovering new data and testing theories. moreover. Critical Legal Studies (CLS). Still. a common theme is that their approaches. See generally Peter H. 39 J. & Gary Peller.

First. CLS scholarship rejects the formalistic premises of the legal process school. is procedural regularity. I now turn to a discussion of this tendency. at 709-23. often conflicting policy goals. at 764-84. and indeterminacy. Id. of whom John Hart Ely is perhaps the exemplar in public law. then. all legal rules and systems are. supra note 142.1" In this view. private interests order their relationships in the shadow of law. different institutions have distinctive roles to play in the legal system.g. contextuality. agencies apply technical expertise. PETER GOODRICH. CLS scholars maintain that rules. and so on. at 765-66. roles derived from the institutions' particular legitimating forms and norms and technocratic capacities. emphasizing instead the importance of law's subjectivity. Id. stress the need to maintain the integrity of the lawmaking process through three principal norms. The institutional competence norm implies differentiation in decisionmaking institutions. and proceduralismmilitate in favor of greater legal complexity. courts interpret and apply the law. 42:1 normative commitment on their part to legal complexity per se. and manipulable. READING THE LAW 16-19 (1986). like all texts. The proceduralism norm implies that legal rights will be surrounded by extensive process safeguards. e. these three norms-institutional competence. juries fiid facts. See. Eskridge & Peller. conditions that putatively simple rules treat as objective and determinate are actually social constructions privileging certain perspectives and groups.'43 Taken seriously and taken together. these institutions should employ utilitarian. . 145. contextual.. purposivism. instrumental rationality methodologies in their interpretation of legal materials. being wholly epiphenomenal. the acid test of the law's integrity. The "new" legal process/judicial policymaking scholars. Purposivism and proceduralism will generate (in my terms) dense. Second. 144. reflects and advances the interests of the already powerful. indeterminate rules. In these related senses.145 The legal system that they constitute. It dictates that legislatures make social policy judgments. constructions that do not exist apart from the interpreter's subjectivity. are 'inherently ambiguous. Finally. and the source of its legitimacy. com- 143.DUKE LAW JOURNAL [Vol. The purposivism norm implies that the law will be interpreted to further multiply. and must be.

because of a few enormous recoveries.19921 LEGAL COMPLEXITY plex. about the extent to which this preference actually affects the law. each of them. complex party structures.J. CHI. 291. Wellington. After all. perhaps the most valuable contribution 'of law and economics to the analysis of law is its systematic attention to the costs of legal complexity as it affects efficiency. must take account of. Schwartz. 87 YALE L.'48 In fact. in order to be efficient. Contributory and Comparative Negligence: A Reappraisal. multiple defenses. and uncertainty itself.146 Efficiency requires minimizing the transaction costs that complexity generates. 273. such as the strategic use of legal procedures. Challenges to Legal Education: The 'Two Cultures' ." scholars' influence may be negligible.147 On the other hand. even the best law and economics scholars sometimes overlook the complexity of efficiency. On the one hand. 697. and be tailored to. any serious pursuit of efficiency (not to speak of equity) will often require complex rules. the mean verdict is much higher than the median. 147. 149. of course. Indeed if the law schools are as increasingly isolated from the legal profession and hence from the rest of society as some have claimed. new damage categories. Peter Huber. Accomplishing this may necessitate a system of multi-factored rules. Junk Science and the Jury. supra note 21. and the mystification induced by the law's complexity can be used to secure a kind of passive acquiescence in-a bogus legitimation of-the system. 1990 U. LEGAL F. and the legal rules. 146. sequential burden-shifting. the goals and constraints relevant to a given policy are likely to be numerous. see POSNER. A controversial example is cited by Peter Huber. and rules that are sensitive to risk preference and insurance market factors. See Harry H. as well as minimizing the efficiency-distorting practices that complexity may facilitate. 703-10 (1978). 148. In their quest for elegantly simple theories. efficient deterrence requires a well-calibrated comparative fault system. Richard Posner's insistence that the common law contributory negligence rule is efficient is perhaps the clearest example of this error. For an example. and so on. Law and economics scholars are more ambivalent about legal complexity than are scholars from the other two schools. who states that "[p]laintiffs' lawyers can profit handsomely from the disorder and inherent unpredictability" that are caused by what Huber calls "the right-field slouch"--an asymmetrical curve of verdicts in which. rentseeking rules. The fact that legal scholars tend to prefer complexity says nothing. No simple rule or set of rules can satisfy all of these constraints. often long causal chains. See Gary T.

LEGAL EDUC. My claim. . 1989) (law reviews in 1985. I also recognize that legal actors who prefer legal complexity on balance may nonetheless feel countervailing pressures for simplicity. This is especially true of bureaucrats and others who must administer law in the face of complexity's costs.DUKE LAW JOURNAL [Vol. but little use of social science). professional. A recent but still unpublished empirical study of the contents of law reviews lends some support to Wellington's view. and are more useful to policymakers and theorists than to practitioners). 6. 812-17 (1981) (noting a growing citation of law reviews. 37 J. 42:1 Nevertheless. all of these stories are plausible. 151. for example. Moreover. which stress how litigation incentives shape law into more complex forms. Given our ignorance about the precise causes of legal complexity.. that a simple rule in a given situation would require too many exceptions or be evaded too easily. It may be. legal actors may not always (or even usually) think in such explicit terms about the choice for or against complexity. Saks. It is that the influentials who systematically shape the legal and political processes have strong personal. Law Journals: Their Shapes and Contents. supra note 149 (mean number of citations of law review articles in judicial opinions rose 59% from 1960 to 1985). my internal rationality story" is neither better than nor inconsistent with either an external rationality account or the significantly narrower internal rationality accounts favored by some law and economics scholars. 773. L. as my external rationality theory"' posits. But see Lawrence Friedman et al. Phenomenon."5 assures that scholars' theories-and their proclivities for complexity-will continue to influence the law's direction. are more dominated by scholars than practitioners. coupled with their growing reliance upon academic scholarship (and a more theoretical scholarship at that). Let me be clear that my explanation accepts the claim that genuine public interest concerns may demand complexity. Michael J. as compared with those in 1960. are more critical of existing doctrine. 33 STAN. is a modest one. Those who would reduce that ignorance need all the help we can get. REV. 152. Saks. and ideological interests in legal complexity that are often effective in securing these interests in the formal law. 1960 and 1985: Address Before the AALS Symposium on Legal Scholarship (Jan. See supra note 102 and accompanying text. then. See supra text accompanying note 101. albeit far-reaching in its implications for reform. Finally. State Supreme Courts: A Century of Style and Citation. courts' growing reliance upon recent law school graduates as judicial clerks. 150. 327 (1987).

Still. and institutionally differentiated? The answer. Eventually. is "no. indeterminate. See STEPHEN BREYER. These costs have led to substantive law changes as well. Trangsrud. Smith. the momentum of objects in motion is limited by friction and gravity. cannot 154. countervailing pressures for simplification begin to affect both the law and how peo153.S. Does this mean that the law will continue to move inexorably along this path. powerful. The response to many regulatory failures-natural gas price controls. tends to increase complexity. however. . In politics. for example-is more extensive regulation rather than reconsideration of the underlying regulatory strategies. competition. REGULATION AND ITS REFORM 240-60 (1982). they become insupportable. perhaps for a long time.g. An example is consumer groups' celebration of the egregiously inefficient tort for assuring "a day in court" which few can actually obtain. system 155. rationalized. Officials often misdiagnose problems and adopt operationally complex reforms.L. JUDICIAL CONFERENCE AD Hoc COMMITTEE ON ASBESTOS LITIGATION. Roger H. E. REPORT TO THE CHIEF JUSTICE OF THE UNITED STATES AND MEMBERS OF THE JUDICIAL CONFERENCE OF THE UNITED STATES 17-22 (1991). legal challenges. these costs are not always immediately apparent. THE LIMITS OF LEGAL COMPLEXITY I have argued that the forces favoring legal complexity are broadly based. The Consolidation of Complex Litigation: A Critical Evaluation of the ALl Proposal.LEGAL COMPLEXITY 19921 IV. at 262-68. supra note 21. Mass Trials in Mass Tort Cases: A Dissent. 1819. 1840-68 (1992). The costs may be rationalized in terms of some transcendent ideal. delays. Richard Epstein. & COM. or reconceive them. 55 The enormous costs of mass tort litigation have led to procedural fixes such as consolidated trials and class actions which only generate new uncertainties. and unfairness. 156. clearly. and systemic. moreover. Lester Brickman.Y The policymaking process. 430 U. An example is courts' insistence that "the cost of protecting a constitutional right justify its total denial. The Asbestos Litigation Crisis: Is There a Need for an Administrative Alternative?. the momentum of triumphant forces is limited by events. can be ignored." or even depicted as benefits. 817. ILL L. 13 CARDOzO L. SCHUCK. 825 (1977). and those benefited by complexity have strong incentives to employ rhetorical and political tactics to conceal. 69 (1989). organizational dynamics. 1989 U. and new ideas." Bounds v. 1 (1990). In the legal system. disguise." In nature. the momentum of complexity is limited by the responses to complexity's rising costs by those who bear them.. then. REv. becoming ever more dense.5 6 Complexity's costs. technical. REV. 10 J. and multiplied for a time. The political momentum and behavioral incentives reach a critical juncture.

at different . Cost bearers often circumvent legal complexity by contracting around it. may have a different view of the relevant costs and benefits of such techniques. however. For a provocative view on this question. at C1. 1992. Instead. Oct. See supra Part IIl. 158. Contract. indicated that in the public sphere of formal law. the forces favoring complexity control the relevant processes and norms and usually overwhelm the simplifiers. the responses to complexity need not be mediated by or effectuated through politics or formal legal changes. 906-08 (1985). J. they can seek to avoid or reduce these costs through informal simplifications of various kinds. Oct. American Admits. We can distinguish at least three informal strategies: contract. rather than leaving such issues to a court and jury. see Owen M. Against Settlement. But see Edwin McDowell. I believe that we have reached that point. A "Neo No-Fault" Contract in Lieu of Tort: Preaccident Guarantees of PostaccidentSettlement Offers. agreeing with others who also stand to benefit from a less complex system to substitute and be bound by a new.. circumvent. simpler regime.. David Field. Hirsch. Fiss. They are devising strategies to . WALL ST. Many producers and consumers of the law now act as if the returns to complexity were diminishing. unilateral simplification. measure of damages. or liability standard. NEW YORK TIMES. REV. Jeffrey O'Connell's various "neo no-fault" proposals employ this approach. 160. 42:1 ple respond to the law."6 So does negotiated rulemaking. 14. In the informal domain. 17. a form of regulatory proceeding in which participants attempt to reach a consensus. at Al (competition for passengers is preventing airlines from maintaining simple structure). 93 YALE L. 159. 1073 (1984). WASH. TIMES. Airfare Plan Fails. 898. 73 CAL. of course. and turn complexity to their own advantage. A recent example in the marketplace is the airline industry's radical simplification of a fare structure that had become so complex that passengers and their agents had great difficulty in dealing with it. Simplifying Their Fares Proves More Difficult than Airlines Expected.157 The reaction against legal complexity proceeds in two domains. Jeffrey O'Connell.. 1992. L. Society.J. . Disputants agree to use arbitration. mediation. There are many familiar examples of this simplification strategy. My discussion of its political economy.159 Contracts specify a particular choice-of-law rule.contain. or other alternative dispute resolution approaches that they regard as less costly than litigation. 157. and withdrawal.DUKE LAW JOURNAL [Vol.. June 4. Pricing Plan Crumbles as Other Airlines Add Innovations.tempos. at A41. 1992. Bridget O'Brien & James S.

Strauss. LAURENCE ROSS." A recent study of the National Highway Traffic Safety Administration found an even more dramatic abandonment of rulemaking in favor of adjudication taking the form of vehicle recalls. Ellickson cites other milieus. Adjudications. The authors attribute this unilateral shift to what I call in- 161.see also Peter L. and Other Sources of Law in an Executive Department: Reflections on the Interior Department's Administration of the Mining Law. 162. 1254-58 (1974). See id. ORDER WITHOUT LAW: How NEIGHBORS SETTLE DISPUTES 257 (1991). analyzed the conditions under which neighbors resolved their disputes according to informal. the agency decided-with mixed results-to abandon rulemaking in favor of policy development largely in the form of adjudications by a special office authorized to grant exceptions to rules. ."' Unilateral Simplification. ELLICKSON. Instead of contracting around complexity. 164.S. simpler set of practices for the more burdensome ones that the law prescribes. §§ 581-590 (Supp. My own study of the Department of Energy found evidence that procedural complexities. extralegal norms that had little to do with the formal rules of property law. Laurence Ross's study of insurance adjustment practices showed that insurers followed adjustment decision patterns that deviated markedly-and invariably in the direction of greater simplicity-from the formal rules and procedures dictated by the tort system and by insurance contracts. Schuck. involving interacting social groups. As these costs rose. 1231. ROBERT C. substituting their own. THE SOCIAL PROCESS OF INSURANCE CLAIMS ADJUSTMENT (1970). cost bearers may act unilaterally.1992] LEGAL COMPLEXITY including an agreement not to challenge the negotiated rule in court. at 194-96. REV. and when the rules otherwise governing the dispute were technically complex. 163. He found-not surprisingly-that they were much more likely to do so when the stakes were small relative to the costs of the legal system. supra note 35. revealingly entitled Order Without Law. in which similar simplification incentives operate. See H.16 Robert Ellickson's recent study of dispute settlement practices utilized by ranchers and farmers in Shasta County. many of them mandated by judicial and executive order. H. at 189-206.C. SETTLED OUT OF COURT:. 74 COLUM. L.163 Other research has unearthed unilateral simplification in the area of public law. had magnified the costs of using informal rulemaking as the mode for energy policymaking. 5 U. Rules. See Negotiated Rulemaking Act of 1990.. III 1991). California. such as water law.

where simplifying-and often arbitrary---caps and defenses have been established" that may engender a complexifying backlash instigated by consumer groups. Schuck ed. unilateral simplification in the face of complexity costs takes the form of partial or complete withdrawal from the field.. Priest. procedural barriers. See MASHAW & HARFST.g. e. For example. 1521 (1987). 96 YALE L. How durable are these formal and informal simplification strategies in the face of the powerful forces favoring legal complexity? Unfortunately.J. At the extreme. MEDICAL MALPRACTICE ON TRIAL 19-69 (1991). delays. Complexity's costs and benefits are remarkably dynamic. In the informal context. 1992). 1991). The Current In. INNOVATION. responding to shifts in technology.16 Withdrawal. while the formal law equilibrium will' be sensitive to such changes only 165.g. THE LIABILITY MAZE: THE IMPACT OF LIABILITY LAW ON SAFETY AND INNOVATION (Peter W. and obstetrical care. WEILER.DUKE LAW JOURNAL (Vol. COMPETITION. 9. this question is very difficult to answer. already faces re-regulation forces. See. A notorious example is the liability insurance availability crisis of the mid-1980s. the equilibrium between complexity and simplicity is likely to move back and forth along the continuum over time. telecommunications law. No.. 42:1 stitutional differentiation-there. public attitudes. vaccines. Litan eds. . Huber & Robert E. George L. and other social vectors. and other costs entailed by having the agency's rules reviewed by courts ignorant of the agency's regulatory tasks.'66 Many other less familiar instances of legal complexity driving out desirable activity could also be cited. 166. at 147-71. AND CONSUMER WELFARE 55-68 (Peter H. surance Crisis and Modem Tort Law. INSTITUTE OF MEDICINE. MEDICAL PROFESSIONAL LIABILITY AND THE DELIVERY OF OBSTETRICAL CARE 38-42 (1989). L. legal simplicity-no less than complexity-also has its cost-driven limits. only recently simplified by a technologically driven deregulation. As these examples suggest.. supra note 13. PAUL C. According to most analysts. politics.. 168. 102-385 (passed Oct. Pub. general aviation. it contributed to a reduction or cessation of risky but socially valuable activity in fields as diverse as after-school sports programs. But this pattern of oscillation does not mean that the formal'and informal regimes will reach identical equilibria. the -uncertainties.67 A similar cycle may occur in medical malpractice law. day care. 1991). Consumer Television Consumer Protection Act of 1992. contraceptive research. For this reason. the equilibrium will be quite sensitive to changes in costs and benefits. TORT LAW AND THE PUBLIC INTERESt. 167. E.

more or less comprehensive schemes may be needed to 169. Some simplifications-deregulation of financial services is an example--cannot be implemented without changes in the law because the complexity is now legally mandated. 162 (codified as amended in scattered sections of 12 U. politicians will tend to strike their balance at a higher complexity level than society in general or cost-bearing groups in particular might wish. even unilateral withdrawal may be illegal unless the regulatory agency formally authorizes it. See. e. Indeed in some regulated industries. related to these market failures. Adverse selection problems.. is that formal. See O'Connell. is "yes.C. however.). 49 U. why not just rely on private incentives to generate informal simplicity where it is warranted? Would that approach leave any problems that simplicity-minded law reformers would still need to address? The answer.S. If my analysis of the political economy of complexity is correct. may suffer from this difficulty.S. may undermine any noncompulsory insurance schemes. This difference between the two regimes helps to explain why simplifications in the formal law regime may not forestall further simplifications in the informal one.C. protects. 170. at 908-10 (acknowledging adverse selection problems of "neo no-fault" approach and proposing ways of addressing them). 89. See 15 U. I think. Glass-Steagall Act of 1933. § 10903(a) (1988) (requiring railroads to obtain approval of Interstate Commerce Commission before discontinuing service along rail lines)." Free rider problems may also discourage even the beneficiaries of simplicity from investing in such a strategy unless the law assures them that others will do likewise.g. and promotes them.S. 48 Stat. 172. Another point.'69 If legal complexity usually takes hold in the formal law regime and cost-justified simplicity finds a niche in the informal. at 153-54 (criticizing Virginia no-fault plan).C. which seek to elide the political obstacles to legislation." Many cost-justified simplicity strategies cannot flourish unless the law formally affirms. supra note 160. Some of Jeffrey O'Connell's imaginative contractual no-fault proposals. for example.Y Even when the complexity is voluntary. See id. some simplifications cannot reap their full advantages unless they are established by law. Even though Virginia lawmakers decided to replace tort law for birth-related neurological injuries with a no-fault system. ch. § 717f(b) (1988) (requiring pipeline operators to obtain approval of Federal Energy Regulatory Commission before abandoning natural gas pipelines or facilities).' In such cases. 171. . any informal simplification strategy short of unilateral withdrawal will probably be illegal.1992] LEGAL COMPLEXITY as they are institutionalized through the political influence of different groups. obstetricians and patients will still search for other strategies that yield even better complexity equilibria.

see Neil K. by discouraging suits or furnishing evidence of knowledge of-a risk. Komesar. apparently derives some of its advantages from lowering administrative costs and what I have called delegitimation costs.DUKE LAW JOURNAL (Vol.2d 968 (Wash. for example-as evidenced by the frequency with which providers unilaterally. however. (B). at 18.Y. See.g. Canada's Health Insurance and Ours: The Real Lessons. The Canadian health insurance system's universality.173 Finally. 22. Wagenblast v. they cannot contract for them with any expectation that their informal simplifications will be upheld by the courts.. as I attempted to do earlier in drawing attention to governance and delegitimation costi. Injuries and Institutions: Tort Reform. 42:1 achieve scale economies or other efficiencies.withdraw from those fields despite the existence of simpler alternatives that would probably leave most or all parties better off.' This is also true for many other fields of activity' 76 -river canoe rentals. 758 P. Under the current state of the law.2d 441 (Cal.g. we can 173. 1963). Odessa Sch. Regents of Univ. even absent these problems. At the first level. e. At least viewed ex ante.'" At the second level of analysis. For discussion of some of the variables bearing on this question. Mashaw. I am indebted to Boris Bittker for pointing this out. See Theodore R. Fall 1990. a limited liability contract may nevertheless have some value to the provider (and thus to the patient)-for example. contracts to simplify will not be entered into unless there is some assurance that the formal law will enforce them. should public-spirited reformers do about legal complexity? Does the preceding analysis provide any analytical leverage for our choices about the amount of complexity that is appropriate to a given legal problem? Improving the balance between complexity and simplicity requires three levels of analysis. 175. 38-47 (1990) (distribution and magnitude of impacts and stakes among actual and potential victims and injurers affects effectiveness of different systems of risk control). of Cal. 176. 1988) (invalidating release form concerning interscholastic sports accidents). Another example of a more comprehensive calcu- . 23.. L.. 65 N. AM. TOWARD LEGAL SIMPLIFICATION: A STRATEGY FOR REFORM What.U. 174. 383 P. See supra Sections II(A). 177. PROSPECT. Medical malpractice law provides an important example. REV. providers and patients would probably both be better off with any of a large number of simpler liability regimes 4 . Even if unenforceable as such. Tunkl v. Tort Theory. V. E. Dist.. we can try to assess complexity's costs and benefits more comprehensively. then. and Beyond. the Big Choices. for example. Marmor & Jerry L.

mimicking. By scrutinizing the informal simplification strategies that groups adopt in order to minimize complexity's costs. Boris Bittker has suggested another criterion for identifying potentially fruitful areas for simplification. and in which the benefits of complexity flow primarily to well-organized groups. The Cost Distribution Principle. I shall call these balancing criteria the cost distribution.19921 LEGAL COMPLEXITY seek general principles to guide our future balancing choices. supra note 73. consumers') interests. and user fee principles. and that it is therefore unlikely to be socially optimal. we can deduce from these cost-minimizing strategies some useful criteria for determining both where to look for promising simplification opportunities and how to implement reform. In these situations. we can gain a clearer understanding of what those costs are." It holds that the complexity of a rule should be tailored to the sophistication and cost-bearing capacities of those who will have to interpret and implement it. more diffuse cost bearers. . More to the point. By studying how complexity costs are distributed. The Audience Principle. poorly organized interests. 178. they will be even less inclined to adopt simpler. as well as of the likely benefits of adopting simpler alternatives. The earlier discussion of the political economy of complexity indicated that the de jure complexity level partly reflects the control that law producers exercise over its development. Again. If law producers can easily externalize complexity costs to diffuse.17 This suggests a criterion that reformers can use to discern when simplification efforts may be fruitful. reformers can identify situations in which these costs are disproportionately borne by groups that face high obstacles to political organization. Call it the "audience principle. complex arrangements may persist for no better reason than that powerful groups are advantaged by them and can free-ride on weaker. An example is consumers who cannot externalize their costs to insurers or to other groups that can organize to protect their own (and indirectly. less costly alternatives. audience. the earlier discussion is instructive. See Gross. Bittker notes that complex tax rules are less problematic if they are addressed to experts who must apply them to uncommon transactions than if they are addressed instead to numerous lay people who must understand and apply them to common lus is Samuel Gross's analysis of the benefits of some complexity-induced inefficiencies.

180. See.. the formal law should not mimic them. 182. and social insurance.g. Ian Ayres & Robert Gertner..g. The Mimicking Principle. Bittker. Indeed. we should consider altering the formal law to mimic the informal simplification strategies. e. protect. moreover. would apply as well to other mass administration areas. at 5.' As a presumptive balancing principle. we need to know how to simplify. such as welfare. Strategic Contractual Inefficiency and the Optimal Choice of Legal Rules. 179. and promote simplification may still be socially beneficial. the focus here is on how complexity affects the democratic morale and self-government in different settings. By studying the ways in which people contract around the formal law. See. One would expect that the public's perception of the law's legitimacy is jeopardized more by the needless and perhaps strategic complexity of. supra note 15. In addition to knowing where to simplify."s the fact that some informal simplification has occurred does not mean that the resulting level is now optimal. 42:1 transactions. Making the formal law and informal practice more congruent is not necessarily superfluous. the regulation of food and drug products. say. 101 YALE L. we can often obtain good information about its inefficiencies. 729 (1992). See supra text accompanying notes 170-76. . would be concerned with effects on small businesses as well as on individuals. while concerned with the plight of individual taxpayers. voter registration.DUKE LAW JOURNAL [Vol. say. if the informal strategies offend public norms or are products of market failures or strategic behavior. Another application of this audience criterion emphasizes the fact that some subjects of legal regulation seem more closely linked to basic democratic values than others. e.I. including its excessive complexities. where simplification might have a large payoff. Having the formal law affirm. In contrast to Bittker's point about the administrative costs of complexity in different settings." 9 Bittker's criterion. that possibility may be an important reason for selecting a particular legal rule based on informal practice. The audience principle. 181. rules on voting and political competition" ° than by the complexity found in. can serve only as a first approximation. This principle. supra note 97 and accompanying text. pension administration. As long as private incentives alone are unlikely to generate the desired complexity level. however.

1" The court may also shift some of a defendant's litigation costs to a plaintiff who rejects a settlement offer and then fails to recover more than the offered amount at trial.g. which would be both unfair and inefficient. For example. even tie most desirable informal practices cannot always be successfully mimicked by the law. § 12111(5)(A) 11 1990). supra note 73. and yet be too vague. See. Gross. . some situations exist in which marginal complexity costs can be accurately calculated and assigned. or too personal to enact as laws. Samuel Gross explains some of the reasons why this is so: Informal norms of behavior suffer from the limitations of their advantages: their relative flexibility. allocate. 184. R.S. and that the effort to do so will seldom be worth its cost. 185. the theory of such exemptions seems to be that imposing the same compliance costs on large and small businesses would put the latter at a competitive disadvantage. Americans with Disabilities Act of 1990. and their responsiveness to interests that legal systems are hard-pressed to regulate-trust. 68. 85' It may be feasible to apply this principle in other situations in which the political economy of complexity can be more closely aligned with the public interest in both achieving an efficient level of complexity and in fairly distributing its costs. 186. too complex. At the third level of analysis in improving the complexitysimplicity balance. at 756. Where feasible.1992] LEGAL COMPLEXITY Unfortunately. etc.direction is exemplified by statutes that exempt from certain regulatory requirements companies below a certain size. the Federal Rules of Civil Procedure authorize the court to assess a party with the costs of any excessive discovery it has occasioned. reputation. P. Still. R.C. CIV. e. or even fully to articulate. and quantify most complexity costs." The User Fee Principle. CIv. FED. FED. A crude effort in this."8 In part. I use the word "special" in recognition of the fact that it is usually very difficult to isolate. (Supp. such norms may operate very well in practice. Because of these features. civility. 37(a)(4). P.. 26(c). lawmakers should seek to tax the special beneficiaries of legal complexity for the special costs associated with those benefits. we can ask whether these criteria point to any specific reforms that seem likely to promote justice by reducing 183. too changeable. 42 U.

See. supra note 168. at 765. attitudinal shifts. ARIz. The Evolution of Useful Life Statutes in-the Products Liability Reform Effort. simpler rules. See Ayres & Gertner.C. ANN. 187. transaction. and delegitimation costs of legal complexity. CONN. ROBERT E. STAT.g. Id. of course. supra note 181. 193. of course. BASIC PROTECTION FOR THE TRAFFIC VICTIM 5-10 (1965) (under such plans." 9 Since transparent rules are more certain and easily administered. that must await more detailed analysis. Whether they can deliver what they promise and whether they are politically feasible are questions. ANN. § 52-577(a) (1991) (product liability actions must be commenced within 10 years after defendant parted with the product). I assume that complexity's beneficiaries will fully exploit any gains from it without our help. 189. e.J. TENN. CODE ANN. Before dismissing them as naive. we should remember that a reform that is politically unattainable now may become more attractive later as the pressure of events."9° as do certain elements of Paul Weiler's recent medical malpractice reform proposal.S.. See.DUKE LAW JOURNAL [Vol. at 114-32. WEILER. 1989 DUKE L. competition. 190. governance. Note. Many simplification reforms focus primarily on reducing uncertainty-related transaction costs. they can be superior even if the rules are not mere defaults and their bargaining costs are high.9 Clear "safe harbor" rules that establish periods of repose in the products liability field" or acceptable practice in the medical malpractice area193 also reduce uncertainty-related transaction costs. however." Commercial law reform proposals. 192 . almost any transparent rule may be an improvement. however. often take this approach. e. STAT. KEETON & JEFFREY O'CONNELL. § 12-551 (1991) (product liability actions must be commenced within 12 years of sale). 188. Van Kirk. 1689. . § 29-28-103(a) (1991) (product liability actions must be commenced within 10 years of date of purchase). The Keeton-O'Connell no-fault auto insurance plan exemplifies this position. and other changes alter its cost-benefit profile."D Some promising simplification proposals do exist.g. the issue is occurrence of accident and injury. 42 U. see also Robert A. § 1320c-6(c) (1988) (exempting physicians and providers from civil liability for actions taken on account of or in compliance with norms applied by peer standards review organizations). GEN. 42:1 the administrative. 191. One approach is to adopt clearer. REV. If bargaining costs are low and the rule is merely a default that the parties may displace by contract. at 737.. Again. not fault).

See supra note 19. Valuing Life and Limb in Tort: Scheduling "Pain and Suffering. at 559-67. Beyond Tort Reform: Developing Better Tools for Assessing Damages for Personal Injury. ON REG. for example.194 policymakers may be less risk averse if they think that innovating in an area that they do govern will not reverberate in others that they do not. which by denying timely compensation to deserving accident victims delegitimates the system."' No-fault systems could reduce the delay and capriciousness of tort litigation. 8 YALE J.. 216-19 (1991). Blumstein et al. supra note 21. consortium.. Schuck.1" Simpler legal rules may also reduce delegitimation costs. 197. at 291. at 923 (providing data indicating that within each of eight "severity levels" of tort claims the highest valuation of damages is generally scores of times higher than the lowest). Peter H. 17 CASE W. Shanker. first-served 194. 908 (1988-89). 171. 196. REv. could limit the great inequities resulting from allowing different juries to award very different damages to victims who are in fact similarly situated. Returning to Fuller's spider web metaphor for the legal system. 198. Strict Tort Theory of Prod- ucts Liability and the Uniform Commercial Code. 8 YALE J. U. however. and other nonpecuniary losses. L. Morris G.19 Similarly. This consideration. Bovbjerg. James F. REV. for example. Schuck. I have proposed that judges with large asbestos caseloads abandon the exceedingly simple first-come. L. Bovbjerg et al. ENTERPRISE RESPONSIBILITY FOR PERSONAL INJURY 199-230 (1991) (considering a similar "scale" approach to the calulation of pain and suffering damages). less institutional dif- ferentiation might reduce the legal indeterminacy that such differentiation tends to spawn. See SCHUCK. See JAMES A. Scheduled Damages and Insurance Contracts for Future Services: A Comment on Blumstein. and Sloan. This "demoralization cost" engendered the demand for sentencing guidelines in the context of criminal law. for example. a more complex rule may be needed to produce a simpler. technical standards issued by regulatory agencies and satisfying certain conditions could be made presumptively binding on juries. 195. more equitable law. 177-86 (1991). supra note 21. supra note 197. not just transaction costs.. Sometimes. PRODUCTS LIABILITY: PROBLEMS AND PROCESS 279-80 (1987). In tort cases. 5 (1965). & AARON D. Scheduled damages.1992] LEGAL COMPLEXITY Simpler legal rules may reduce governance costs. RES. ON REG. TWERSKI. See AMERICAN LAW INSTITUTE. .9 6 Jury discretion could be constrained by using scheduled damages to compensate pain and suffering. should counsel courts against extending tort principles into the realm governed largely by private contractual arrangements under the Uniform Commercial Code. See supra note 80 and accompanying text. HENDERSON. See Bovbjerg et al. For example." 83 Nw. JR. 213. Randall R.

should be broad indeed. For a summary of such proposals."' Pruning away the needlessly. See S.. Ackerman & Richard B. (1991) (proposed National Voter Registration Act of 1991. 71 MARQ. Environmental protection furnishes many opportunities for reforms of this kind. GEN.. 102d Cong. see also Peverill Squire et al. international trade. 250. Other proposals fall under the rubrics of incentive-based regulation. 204. 2' Financial services. Reforming Environmental Law. See supra note 97 and accompanying text. see Ronald Cass. after applying medical criteria. 203. See.' " Although the proposals I have just mentioned are directed primarily at the tort system. voter registration-an inconvenient. 4366. supra note 21. time-consuming process in many jurisdictions-could be facilitated by simply linking it to applications for a driver's license. and biotechnology are attractive candidates for such reform.g.. REV. 2d Sess. WOLFINGER & STEVEN J. the critique of legal complexity extends far beyond this area to all areas of public policy.R. 449 (1988). 42:1 rule for scheduling trials in favor of a more complex rule that would sequence claims according to the seriousness of plaintiffs' present injuries. then. 102d Cong. See Bruce A. 200. STAT. 57-58 (1987) (proposing simplified registration procedure). . indeed cynically. H. CONN. energy. under which voter registration would take place during driver's license application and renewal). The focus of simplification reformers. POL.DUKE LAW JOURNAL [Vol. also known as the "Motor Voter" bill. Privatization: Politics. 19a-279(b) (1990) (requiring state to give drivers an opportunity to indicate on driver's licenses whether they wish to donate organs in event of fatal accident). 81 AM. complex signature and petition requirements from New York's election law would prevent partisans from exploiting its complexity to foster costly litigation and deny serious candidates access to the ballot. WHO VOTES? 80-88 (1980). Law and Theory. e.' A similar technique can be used to encourage people to consent to donating their organs in the event of a fatal accident. Sci. For example. RAYMOND E. 1st Sess. See Schuck. L.2M A large number of far-reaching simplification reform proposals in the area of public law involve deregulation.. (1992) (providing for voter registration at time of driver's license application and renewal). Residential Mobility and Voter Turnout. including emissions trading and pollution taxes. 45. ROSENSTONE.' 4 Performance-based management is 199. 201. REV. Professor Raymond Wolfinger has studied the effects of complex voter registration procedures and argued that simpler procedures would strengthen the legitimacy of the electoral process without having significant partisan effects. §§ 14-42(b). Stewart. their goal should be to increase the supply of socially valued activities by reducing the complexity-related costs of performing them. 202. At its most general level.

208. as in public education and health care. MARKETS & AMERICA'S SCHOOLS 199-201 (1990). sunset reviews. . legislative revision commissions. There is abundant evidence of other societies' inability to manage it. work in precisely the way they were envisioned. CONCLUSION An analysis of legal complexity. REV. and administrators of law must think more carefully about the complex rules and systems that they have tended to prefer. then. is not that they are perfect. 5 Still other proposals involve a change in the mix of market and regulatory forces. What makes such reforms at least plausible. 37 STAN. PAUL STARR. MOE. See JOHN E. the earlier faith that improvements in information technology and other management techniques would enable bureaucracies to keep pace with complexity in these areas seems hopelessly fatuous. holds important lessons for lawmakers. See generally DAVID OSBORNE & TED GAEBLER. from the strong privatization movements in virtually all Western European nations to the collapse of centrally administered economies in Eastern Europe and other parts of the world. REINVENTING GOVERNMENT: How THE ENTREPRENEURIAL SPIRIT Is TRANSFORMING THE PUBLIC SECTORS (1992) 206. of course. Several techniques now in use for other purposes-for example. THE LOGIC OF HEALTH CARE REFORM (forth- coming 1992). Perhaps a more systematic view of the causes and consequences of complexity may bring them to different conclusions about its merits in one legal realm or another. 1341-51 (1985). L. 207."° More generally. and regulatory and small business impact statements-might serve as fruitful models. rationalizers. If anything. 1333. moreover. POLITICS. complexity's resistance to centralized control seems to have increased over time.' In retrospect. are neither easy to establish and maintain nor unproblematic in their operation. and scholars.19921 LEGAL COMPLEXITY an analogous reform within bureaucracies. See supra text accompanying notes 39-40. then. policymakers should institutionalize a periodic review designed to purge laws of their unwanted encrusted complexities.n Efficient markets. but rather the fact that centrally coordinating innumerable individual and firm decisions fairly and efficiently has proved to be an unmanageably complex task. VI. CHUBB & TERRY M. The professional producers. Few legal simplification efforts. 205. citizens. as the experiences with airline deregulation and sentencing guidelines demonstrate.

See.A.. 30. 92 Stat. L.).S. L. Farina. L.21° the Paperwork Reduction Act. No. DOING AWAY WITI4 PERSONAL INJURY LAW (1989). 94 Stat. REV. and commitment to understanding complexity equip us to conduct relatively sophisticated. e. 96-511. 96-296.). however. 94-210.g. RICHARD A. We scholars also have a crucial role to play in the campaign to domesticate legal complexity. POSNER. 209. 458-61 (1989). 42:1 Given the political economy of complexity. 94 Stat. 211. THE FEDERAL COURTS: CRISIS AND REFORM (1985). broad-gauged studies of its effects in different legal settings.S. 90 Stat. Inc. See.C. and refining them as their imperfections become manifest. 42. see Cynthia R.C. 20.S.S. 213.S. Chevron U. .. Our professional detachment. No. 89 COLUM. and of the merits of proposed reforms that seek to strike a superior balance. No. 49 U. Statutory Interpretation and the Balance of Power in" the Administrative State.S. 467 U. e. 21 3' But we are educators as well as scholars.g. L.C.DUKE LAW JOURNAL [Vol. 793 (codified . 837 (1984). law producers are unlikely to take this more systematic view on their own unless citizens insist upon it by rewarding law producers when they devise worthy simplifications and penalizing them when they embed excessive complexity in the law. 31 (codified as amended primarily in 45. Railroad Revitalization and Regulatory Reform Act of 1976.C.).g. reminding anyone who will listen about the elusive virtues of simplicity in law. SUGARMAN. 44 U. analytic training. 452. Inc. Pub. e. As we learn more about legal complexity's consequences.. We know that the case for simplicity in law can occasionally carry the day politically. in scattered provisions of 49 U. 95-621. of the strategies that people adopt in its shadow. STEPHEN D.). Pub. v. Motor Carrier Act of 1980. The deregulation of surface transportation 9 and energy production.. Natural Gas Policy Act of 1978. For an analysis of Chevron. flawed though they surely are in some respects. See. L. Pub. Pub. 210. 212. No. 2812 (codified as amended in 5. Sustaining these simplification victories over time. will be more difficult. we should infuse that learning into the political economy of complexity.' and the "bubble" approach to pollution control21 2 are examples. 3350 (codified in scattered provisions of 15 U. Natural Resources Defense Council.